Skip to main content

Full text of "The Story Of Musical Instruments"

See other formats

7$7 S39s 950010 


Story of musical instruments 

kansas city 

public library 

Kansas city, missouri 

Books will be issued only 

on presentation of library card. 
Please report lost cards and 

change of residence promptly. 
Card holders are responsible for 

all books, records, films, pictures 
or other library materials 
checked out on their cards. 

3 1148 00115 9011 

DEC 2 3 1375 
IVIAI MAH B i vq(- ' 

JUL si 1977 


^j 1 G 'j 
JUN 20 1978' 

LB2 1979 

MAR 5 1980 


, '-.j'p ?. \ 1981 


MM AU&-2-6-HH 

IWU. MAR? 1992 




199 , 


Musical Instruments 




Doubleday, Doran & Company, Inc. 

PR TNT" EB'ATfrHK Cowttti Life Press, GARDEN CITY, N. Y,, if. 8. A, 







for the use of most of the photographs and drawings 
shown in this book, and to Dr Robert Young, physicist of 
C. G. Conn Ltd., for reading and criticizing the manuscript 
of the final chapter, "How Music Is Made." 

Grateful acknowledgment is made to the following musi- 
cians who read in manuscript the respective chapters on 
their individual instruments: A, Hilsberg, solo violinist, 
Philadelphia Orchestra; William Kincaid, solo flutist, Phila- 
delphia Orchestra; Thomas J. Byrne, oboist and English horn 
player, Detroit Symphony Orchestra; Clarence Warmelin, 
Chicago teacher and lecturer on woodwinds, formerly solo 
clarinetist, Minneapolis Symphony Orchestra; Lucien Cail- 
liet, solo saxophonist, bass clarinetist and arranger, Philadel- 
phia Orchestra; Harry Glantz, solo trumpeter, Philharmonic- 
Symphony Society of New York; Simone Mantia, solo 
trombonist, Metropolitan Opera Orchestra; Max Pottag, 
French horn, Chicago Symphony Orchestra; Philip Dona- 
telli, principal bass tuba, Philadelphia Orchestra; and 
William F. Ludwig, former percussionist, Chicago Grand 
Opera Orchestra and Chicago Symphony Orchestra, 



Full responsibility for the contents and treatment of the 
various chapters is assumed by the author, but these eminent 
authorities on their instruments have helped make the book 
more accurate, and they have contributed many valuable 
suggestions and comments. 






Ingredients and Quantities I 


But Opera Blazed the Trail 18 


Aristocrats of the Orchestra 3$ 


A Story of Great Inventions 57 


The Expatriated Orientals 86 


The Ugly Ducklings m 


The Scapegoats of Jazz 135 


Sixty Centuries of Rivalry *5* 


Graduates of the Chase l8 5 


The Dr Jekyll and Mr Hyde of the Musical 

World 2 8 





The Story of Large Families 232 


Tinkle! Jingle! Boom! Crash! 252 


The Story of Steel Wire 271 

XIV How Music Is MADE 

The Raw Material 296 

INDEX * 359 

[ mii ] 

Halftone Illustrations 


Range chart for band and orchestra instruments . . 10 

Development of violin family 42 

Stradivarius violin, front and back 43 

Development of the flute 58 

Two views of modern bassoon reed, old-type bassoon 

reed, modern oboe reed, old-type oboe reed ... 90 

Development of the oboe 91 

Derivatives of the oboe 106 

The clarinet mouthpiece and the primitive chalumeau 

mouthpiece from which it developed 122 

Development of the clarinet 123 

Parts of the alto saxophone 138 

Development of the saxophone 139 

Trumpet, cornet and fluegelhorn mouthpieces . . . 162 

Development of the cornet and trumpet .... 163 

Parts of the cornet and trumpet 170 

Cross sections, showing the difference between the 
French-horn mouthpiece and the mellophone mouth- 
piece 194 



The ancestor of the French horn the Hebrew shofar . 194 

Modern double French horn in F and Bb .... 195 

Development of the trombone 210 

Cross sections of the trumpet and trombone mouth* 

pieces 210 

Bass trombone 211 

Valve trombone 211 

Development of the tuba family 234 

Instruments of a band of Civil War days . 235 

Mouthpiece of saxhorn and mouthpiece of modern tuba 250 

American Revolutionary War drum 250 

Drums of the type carried by the Moors in the eighth 

and the Crusaders in the thirteenth centuries . . 251 

Line Cut Illustrations 


Showing where the musical scale lies within the limits 

of the audible scale 3 

Conventional arrangement of the symphony orchestra 12 
Seating arrangement of the modern concert band . 13 
How fork fingering was used to play the semitones on 
flutes and other woodwinds before keys were in- 
vented 66 

How Boehm was enabled to open and close fourteen 

holes with nine fingers, by the use of the ring key . 78 
Showing how Denner solved the problem of the chalu- 

meau scale 119 

Trumpets vs. cornets for popular favor . 152 

Showing the comparative lengths of the three valve 

slides of the cornet and trumpet 174 

The bugle scale 175 

Section of the trombone scale 210 

Showing the positions of the trombone slides as used 

to bridge the gaps in the open-note scale . . . 227 
Notes playable and not playable on the tenor and bass 
trombones 228 




Lengths of various instruments and lengths of their 
tone waves 36 

Vibrating a pin, stuck in wood, to create a musical note 3 1 1 

Compression and rarefaction of molecules in a 
plete tone wave 312 

Illustration of the principle of resonance . . . .316 

Drawing to represent difference between closed pipe 
and open pipe, sounding same fundamental note . 321 

Showing how the tone wave divides and produces the 
open tones of the natural harmonic scale. . . . 329 

The natural harmonic scale, produced when the tone 
wave of a horn divides into smaller sections . . . 331 

Diagram showing how waves break up into smaller 
waves to form various notes of the harmonic scale . 333 

Illustrating that the characteristic in the vowel sound 
in "three" is often filtered out when transmitted 
over the telephone and comes out at the receiver 
sounding like "two" 341 

Oscillograph records 348-54 

[ xii ] 


Musical Instruments 


Recipe for a Symphony 


THE GREAT PIPE ORGANS are marvels for variety of tonal 
coloring. The pipe organ manual has more "stops" 
to pull and more gadgets to work than the dash of an air- 
plane or the control room of a submarine. 

But the orchestra excels even the pipe organ in the variety 
of beauty of its tonal coloring and in the amazing wealth of 
its musical effects. The orchestra conductor can "pull stops" 
on the orchestra that are the envy of the organist and the 
despair of the organ builder. 

In the lower regions he can call out the ominous thunder 
of the tympani, the sonorous boom of the tuba, the Plutonic 
mumble of the bassoon, the dark, muffled zoom of the string 
bass, or the sepulchral moaning of the bass clarinet. 

To carry the melody or tell the story of the composition, 
the conductor can call upon the versatile virtuoso violin, 
the coloratura-soprano flute, the lyric-soprano oboe, the 
dramatic-soprano clarinet or the martial trumpet and piccolo. 

For middle voices he can choose the tenor trombone or 
viola, the English horn or alto clarinet, the French horn or 

cello. To beat a rhythm or set a tempo or punctuate a phrase, 
the conductor may choose among the many varieties of 
drums and bells and chimes, or call upon the strings to play 
pizzicato or the trumpets to play staccato. 

The high harmonics of the strings can picture the ethereal 
realms of heaven, or the brass and the battery can blast the 
^hearing with the echoes of hell. The flutes and oboes can 
paint a Corot scene of pastoral contentment, the trumpets 
and trombones can fan our warring spirit to white heat, 
the French horns can call from Alpine peak to Alpine peak, 
or the bassoon can perform the antics of the clown and pic- 
ture the zigzag, uncertain course of the drunkard* The 
clarinets can dance the swift, sgrightly folk dance, the drums 
and piccolo can beat the cadence of marching armies. Or the 
slow, measured beat of the tympani and the low, muffled 
swish of the string bass can pace the funeral march* There is 
nothing, apparently, beyond the capacity of this greatest of 
all musical instruments the symphony orchestra. 

The symphony is composed of about a hundred instru- 
ments and has a range of about a hundred semitones. Since 
the sound limits of the human ear are about 125 semitones, 
the symphony orchestra utilizes about four fifths of the range 
of human hearing. 

The lower threshold of hearing is usually set at sixteen 
vibrations per second, and while the pipe organ sometimes 
uses this tone, four octaves below Middle C, the lowest note 
used in the symphony is the Bb in the fourth octave below 
Middle C, sounded by the giant contrabass tuba and having 
twenty-nine and a fraction vibrations per second. 

Composers pretty generally have agreed that the ten 
semitones in the scale below this Bb cease to be musical 



21,094.4 cycles per second 
Wave .647 inch long 

4185.6 cycles per second 
Wave 3.26 inches long 

A 440 cycles per second 
Wave 31.02 inches long 

C (Middle C) 261.6 cycles per second 
Wave 52.17 inches long 

Rb 29.2 cycles per second 
4 Wave 468.4 inches long 

C 16.4 cycles per second 
4 Wave 834.7 inches long 


Showing where the musical scale lies ^t 
cate the notes referred to in the notations on the right. 

and can better be dispensed with. Not only do they lose 
their musical quality but they literally fall apart. Instead of 
sounding as a smooth, continuous note, fewer than twenty- 
nine vibrations per second tend to become separated into 
individual, distinct pulsations, just as in moving pictures 
fewer than twenty-four pictures per second cease to appear 
as a smooth, continuous movement of events and begin to 
blink and separate into individual, distinct pictures. 

The upper threshold of hearing is about twenty thousand 
vibrations per second. There are some individuals who can 
hear sound of higher rates of vibration than others, and 
women generally can hear sound of higher pitch than men, 
but this is ordinarily conceded to be the extreme limit for 
anybody. The top note of the orchestra, however, has only 
4185.6 vibrations per second. This note is the C in the fourth 
octave above Middle C and is delegated to the tiny piccolo. 
The upper limits of hearing extend about two and a half 
octaves above this C, but these notes are so thin and piercing 
that they give no pleasure to the ear and are not scored for 
the orchestra. This is not to say they are not to be heard in 
the orchestra, for they do actually exist as high, delicate 
harmonics and have much to do with the tonal coloring of 
the music, but our ears do not hear them as notes of definite 

Well-trained human voices have a range of about thirty 
semitones, or two and a half octaves, Sembrich is said to 
have had a range of thirty-four semitones, from G below the 
treble clef to F above high C, or nearly three octaves, but 
this is exceptional. Curiously enough, most instruments of 
the orchestra are no more gifted in range than the human 
voice, most of them having a range of about two and a 


half octaves. Among such are the oboe, English horn, trum- 
pet, trombone, baritone horn, saxophone and string bass. 
A few of them, such as the bassoon, clarinets and bass tuba, 
exceed three octaves, and only two, the violin and cello, can 
encompass four octaves. The piano comes nearest to equaling 
the range of the symphony orchestra, having eighty-eight 
semitones, a few over seven octaves. The harp is runner-up, 
with a range of about seventy-eight semitones, or over six 

The symphony orchestra was a long time in the making, 
and its parts were collected from many strange lands. Snake 
charmers from the Orient contributed the oboe. Ancient 
Greeks before Homer developed the primitive clarinet. 
Horns were probably first used in the religious ceremonies 
of the God-fearing Israelites. Conquering and militaristic 
Rome brought trumpets to a high point of favor, while 
fifty generations later this same land took pride in its superb 
violins. At a late date tubas sprang from Europe, and still 
later the saxophone and sarrusophone. Africa is famed for 
its drums and Greece for its pipes of Pan, or flutes; but no 
one country can claim credit for these two families of in- 
struments, because there is hardly a spot on the globe where 
these instruments have not flourished, and every primitive 
people of today makes and uses drums and flutes in some 
form or other. 

Makers of the instruments of the band and orchestra 
search the world for their materials. The collecting of these 
materials furnishes enough romance to fill a volume. The 
mellow clarinet note in today's concert reminds us of great 
labors and sacrifices of native blacks in the tropical wastes 
of South Africa. The weird sound from the temple block 


echoes the woodsman's ax in the depth of the redwood forest 
of China. The plaintive song from the reedy oboe was made 
possible by the dry soil and bright sun of the Mediterranean 
beach in southern France. 

Without reeds, we would have no woodwind choir, no 
clarinets, oboes, English horns, bassoons or saxophones* 
And if nature had not chosen to throw together in one little 
spot in southern France a peculiar combination of dry top- 
soil, a subsoil moistened by the salty seepage from the blue 
Mediterranean Sea, a unique mixture of organic substances 
to nourish the roots, and a warm sun from a pleasant sky to 
bathe the leaves, we would have no reeds. Cane grows in 
many spots on the globe, but not as it grows in a small area 
in France along the Mediterranean Sea known as the Var 
district, near Marseilles. The finest reed cane in the world 
comes from here, for the soil and climate seem to have con- 
spired together to produce an ideal material for setting into 
vibration the column of air in the woodwind instruments* 
If the climate were warmer and more moist, the cane would 
grow too fast and the reed would be too porous. If the climate 
were not so warm and the soil were drier, the cane would 
grow more slowly and the reed would be too hard. Nor has 
man been able to add anything to this ideal combination, for 
the best cane grows wild and in its natural state; cultivated 
cane is inferior. 

Elaborate pains are taken in preparing the cane after it is 
grown. For three years the cane is carefully cured one year 
in the dry and shade, then six months in the sun, with regular 
periods later on in the sun as the cutting, trimming and 
sorting process goes on. Finally a small piece of cane about 
the size of a stick of chewing gum is produced, but there Is 


nothing else in the whole world which can equal it for sound- 
ing the characteristic tone of the woodwinds. 

If you should suddenly decide one day you were going to 
make the finest violin bow it is possible to produce, you 
would have to take a ship to the port of Pernambuco in 
Brazil. Then you would have to make a hard and long jour- 
ney to the interior of this great country. There, after diligent 
search, you would find growing in the hard and rocky ground 
\ a tree called the Brazilwood tree. You would then select a 
small tree, cut away the outside sapwood, and finally come 
to a small heart, dark red in color. This is the Pernambuco 
wood known to commerce. It was selected by Tourte, the 
great French bowmaker of the eighteenth century, as the 
best material for making violin bows, and nothing finer has 
ever been discovered. It has just the right weight for balance, 
the right grain structure for retaining its shape, and the right 
resiliency for the utmost in bow technique. 

Early in the development of the fine violins of Italy, the 
Amati and Stradivari craftsmen found there was nothing 
like the giant Norway spruce or Swiss pine for a violin top. 
These great trees grew up to heaven for a hundred to a 
hundred and fifty feet, and their grain was even and straight 
as parallel beams of light. This wood fulfilled the needs for 
a material of great elasticity but light weight. They didn't 
know then, but scientists have since found out, that sound 
travels faster through this wood than any other, attaining a 
velocity through the grain lengthwise of fifteen thousand feet 
per second, a velocity nearly equal to that found in steel. 
They didn't have technical proof of this, but their trained 
ears told them that this sprucewood gave the best results. 
They cut the logs through on the quarter and sawed through 

the center. Then the center edges were glued together. This 
gave them identical grain structure from the center to both 
outside edges, the grain being so uniformly even and regular. 
After three centuries of violin making, nothing has been found 
which surpasses this wood for violin tops, and so makers 
today still fell the giant spruce for this part of the violin. 

Several tropical woods have been used for making the 
bodies of the woodwind instruments. The picturesque cocoa- 
nut tree was one of them. The best wood for this purpose 
comes from the West Indies and Central America, It grows 
best in the sandy soil along the sea, or not far inland. In 
contrast to Pernambuco wood, only old trees are used. The 
heart is cut out of the lower portion of the tree trunk. This 
wood is brown in color, heavy in weight, hard to cut, but 
can be polished to an almost metallic luster. It is known 
commercially as cocuswood, and many fine woodwind in- 
struments have been made from it. Some instruments are 
still made from cocuswood, but it has generally been aban- 
doned, for it contains a resin which causes skin poisoning. 

Another wood used for making clarinets, flutes and oboes 
was boxwood. True boxwood comes from Venezuela, but 
most of the boxwood used in musical instruments came from 
the West Indies. It is very tough and has an extremely fine 
texture, but it has one serious defect it warps. This defect 
was not so serious when the key system of musical instru- 
ments was limited to a half-dozen single keys, but when 
several keys were mounted on a single long hinge, slight 
warping caused the hinge to bind, and boxwood had to be 

To the rescue came grenadilla wood, known also as Mo- 
zambique ebony and as African blackwood. This wood is 


cut from the arid wastes of Mozambique, South Africa, or 
the huge island across the channel to the east, known as 
Madagascar. Great hardships are experienced by the sweat- 
ing black natives who bring this valuable wood out of the 
desert lands. Great sand storms come up and bury their 
camps and even the trees they come to cut and haul to the 
coast. Such trees grow slowly, and the wood is very close- 
textured, very heavy, very hard. When first cut it is a 
beautiful dark purple in color, but becomes black when 
cured and oiled. When worked in a lathe the wood will 
almost take the edge off a steel tool, it is so hard, and the 
fine dust resulting is like pepper in the nostrils. After five 
to ten years of curing in unheated wood lofts, the grain 
becomes so fixed, it is nearest to being crackproof of any 
wood. So far, nothing has been found which quite equals it 
for making the bodies of the finest woodwinds. 

There are many other odd and interesting woods which 
will be found in every symphony orchestra. Drumsticks 
will be found made of snakewood from Dutch Guiana, a 
reddish-brown wood with spots in color from brown to black, 
similar in markings to that of a snake. The fingerboards, 
pegs and tailpieces of violins are made from African ebony, 
brought ail the way from the tropical forests of Africa. 
The bars of xylophones and marimbas and often the castanets 
are made of rosewood from Brazil or Central America. 
This wood ranges in color from reddish brown to deep purple 
or black, with streaks of purple through it, and it gets its 
name from its fragrance. It is ideal wood for the purpose, 
since it is hard, dense, takes a high polish and produces a 
brilliant, resonant tone when struck. Then there is mahogany 
for drum shells, hickory for drumsticks and drum hoops, 


walnut for tambourines, maple for violin backs and bassoons, 
and basswood for "cases. 

Some people find less music in the orchestra when they 
learn that the death of from two to three sheep is required 
before one violin can be equipped with strings. It doesn't 
seem necessary, but such is the case, and such is the price of 
great music. So-called "catgut" used on violins is made from 
the intestines of sheep. The average length of the sheep in- 
testine is twenty-four feet, and it requires from ten to twelve 
half-intestines to spin a string only four one hundred ths of 
an inch in diameter, since only the fine, soft, submucous 
membrane is used. A set of four violin strings is eighty-eight 
inches long on the average, and into a set of four violin strings 
go the intestines of from two to three sheep. 

Even old Dobbin contributes his bit, for what would a 
violin bow be without horsehair? Makers today use 150 hairs 
about twenty-eight inches long in a standard bow, although 
Tourte, who created the violin bow of modern times, used 
slightly fewer hairs. Then the pig comes in for his share, for 
tom-tom heads are usually of pigskin. All fine drumheads are 
of calfskin, the thinner and finest heads coming from the 
skins of "slunks" or unborn calves. Many different materials 
are used for mallets of bass drums, tympani, marimbas and 
bells. Among them are yarn, felt, soft rubber, hard rubber, 
rawhide, lamb's wool and pyralin. 

Copper is the most important metal used in the construc- 
tion of band and orchestra instruments. The big tympani or 
kettledrums are drawn from one piece of copper sheet into 
the half-sphere bowl. Copper is also the principal ingredient 
of brass used in the bells of brass instruments, about seventy 
parts of copper being combined with about thirty parts of 



zinc to form brass. Valves, keys, braces and other parts of 
the brass instruments are made from a brass alloy incorporat- 
ing in addition small quantities of tin, aluminum, lead or 
other metals, depending upon what is expected of the part. 
Metal clarinets and flutes are often made of what is called 
nickel silver, an alloy of nickel and other metals. Nickel, 
chromium, silver and gold are used in plating various parts 
of many instruments used in the band and orchestra. 

From the forests, from the foundries, from deep mines and 
across great seas, from desert wastes and tropical jungles 
come, in a real and true sense, the great music of the sym- 

If asked to give a recipe for making a symphony orchestra, 
the historian might start out: Take some oboes from the 
Orient, some clarinets from Greece and some horns from 
Palestine. The man of commerce might start out: Take some 
cane from the Var district of France, some Pernambuco wood 
from Brazil, and some grenadilla wood from Mozambique, 
South Africa. 

Another and better way would be to say : To form a sym- 
phony orchestra of about one hundred pieces, make a mixture 
of about 40 per cent violins and violas, about 20 per cent 
cellos and basses, with just a sprinkling of a harp or two. 
Make another mixture of woodwinds composed of flutes, 
single reeds and double reeds, up to about 15 per cent. Make 
an equal quantity of brass mixture, composed of trumpets, 
trombones, horns and tubas. Pour all three mixtures together 
and add about 6 per cent tympani, drums, bells and traps. 

The recipe for a modern concert band would be something 
like this: Make a mixture of about 38 per cent clarinets 
soprano, alto and bass. Add to it about 10 per cent other 

woodwinds oboe, English horn, bassoon, flute. Then make 
a mixture of about 38 per cent brass, including cornets, 
trumpets, trombones, horns, baritone horns and bass tubas. 
Pour these two mixtures together, stirring in at the same 

Conventional arrangement of the symphony orchestra. This is 
often varied to allow the cellos to play on the right of the conduc- 
tor. The trumpets are also sometimes to the left of the trombones 
and the tuba to the right of the trombones. Often the harps are 
placed on the right side of the stage. 

time about 8 per cent saxophones. Then add about 6 per cent 
tympani, drums, bells and traps. 

Examination of the instrumentation of the great symphony 
orchestras of America shows but slight variation from the 
recipe given above. There may be slight differences within 
the choirs themselves, but the general balance among the 
string, woodwind and brass choirs and the percussion section 
is substantially uniform. Comparison of the instrumentations 
pf the various great symphonies follows; 


First Violins 

Second Violins 
f\ Violas 
\J Cellos 


Flutes and Piccolos 

Oboes and English Horns 


Bass Clarinet 
































o is 












.jj j* 

















Seating arrangement of the modern concert band. Some arrange- 
ments provide for all the woodwinds on the left and the brass on the 
right, but the tendency is toward this arrangement, sometimes 
called the symphonic arrangement. 


The instrumentation of bands is not so well formulated, 
but there is beginning to be greater uniformity among the 
better bands. Here also there is fairly established balance 
among the woodwind and brass choirs and the percussion 
section, even though there is considerable variation within 
these choirs. Below is a table, showing the instrumentation 
of several well-known American bands. 













Flutes and Piccolos 






Oboes and English Horns 












Alto Clarinets 






Bass Clarinets 

















Cornets and Trumpets 







- 8 














7 ' 





















String Basses 








Total 86 76 76 61 91 

Berlioz, the famous conductor and instrumentation au- 
thority of the nineteenth century, should have been an 
American, because he had the American instinct for "bigger 
and better." After discussing all the various instruments of 
the orchestra, in his great work called Modern Instrumenta- 
tion and Orchestration, he closes the book with a suggested 


instrumentation for a giant orchestra of 467 instrumentalists, 
which he believed would be capable of great and new artistic 
achievements. Such an orchestra, he says, would be made up 
as follows: 

1 20 violins 
40 violas 
45 violoncellos 
1 8 double-basses with 3 strings 

15 double-basses with 4 strings 
4 octo-basses 

6 large flutes 

4 third flutes 

2 octave piccolo flutes 

2 piccolo flutes 

6 hautboys (oboes) 

6 corni Inglesi (English horns) 

5 saxophones 

4 bassoons-quinte 
12 bassoons 
4 small clarinets (Eb) 
8 clarinets 

3 bass clarinets 

16 horns 

8 trumpets 

6 cornets 

4 alto trombones 

6 tenor trombones 

2 great bass trombones 

1 ophicleide in C 

2 ophicleides in Bb 

2 bass tubas 
30 harps 

30 pianos 

1 very low organ 

8 pairs of kettledrums 
6 drums (snare) 

3 long drums 

4 pairs of cymbals 
6 triangles 

6 sets of bells 
12 pairs of ancient cymbals 

2 very low great bells 
2 gongs 

4 pavilion dhinois 

467 instrumentalists* 

Johann Strauss III once conducted in Vienna what is 
probably the largest symphony orchestra ever to play under 
one baton. In this colossal orchestra were one thousand 
players, or ten orchestras in one. But the laurels for the 

*A Treatise on Modern Instrumentation and Orchestration, by Hector Berlioz, 
translated into English by M. C. Clarke, is the source of this interesting addition. 
The skeptical reader will have found before referring to this note that the total 
467 is not the sum of the figures itemized. The actual total is 465. Reference to 
the original French by Berlioz reveals the figure 458 as the total, which when 
added becomes 456! Music in its strictest sense is mathematical, but it does not 
seem to follow that musicians can add. 

world's largest band ever to play under one baton go to 
America. This monster band played during the 1933 National 
High School Band Contest in Evanston, Illinois, held in 
connection with the Chicago Century of Progress Exposition, 
Over five thousand bandsmen from over sixty bands took 
part, and the concert of six selections was broadcast over a 
national hook-up. 

The formation of the band is graphically described by 
Robert L. Shepherd, editor of The School Musician, an eye- 
and-ear witness : 

"From his tall conductor's stand, Mr Bainum now 
directed the formation of the greatest massed band ever 
assembled. Using the field amplifier he first brought the 
color bearers forward. Next the enormous drum section 
of sixty basses and two hundred and forty snare drums 
were formed immediately behind the colors in the center 
of the field. Then, in order, and in relatively fine formation, 
the first, second, third, fourth and fifth horizontal rows of 
bands, as they were positioned in the stand, marched 
down and out across the field, falling naturally into their 
places in the great spotted mass of brilliant color and 
glistening instruments. It was a thrilling spectacle, a 
pageantry that would have made a great genius like Rich- 
ard Wagner fairly weep for the beauty of it. Words seem 
futile in an attempt to paint the picture. It was a dazzling 
panorama. 3 ' 

The director passed the beat to the three hundred drums 
immediately in front of him, and they in turn boomed out 
the tempo to their fellow bandsmen who surrounded them* 
In this manner the five thousand musicians played in re* 


markable unity and with surprising precision. Fourteen 
hundred clarinets sang together the stirring strains of Sousa's 
"Stars and Stripes Forever" and five other marches, while 
seven hundred cornets and trumpets and four hundred 
trombones chimed in. The harmony was swelled by four 
hundred French horns, two hundred baritones, and was col- 
ored by four hundred saxophones and several hundred mis- 
cellaneous woodwinds. Supporting the whole diapason of 
melody, three hundred and fifty giant bass tubas and sousa- 
phones poured out a flood of sonorous chords. It was a musical 
spectacle such as the world had never seenjbefore. 


How the Orchestra Grew 


WE MUST READ three thousand years of recorded history 
before we learn of the music of the Greeks, and another 
three thousand years before we learn of the beginnings of 
opera in Italy, but in the next three hundred years the modern 
orchestra was conceived and developed, with all its new 
ideas of harmony, instrumentation, orchestration and instru- 
ments. These three hundred years are by far the most amaz- 
ing in the history of music. More progress was made in these 
three centuries than in the sixty centuries which preceded 

Although the wonderful Greek mind anticipated most of 
today's arts and learning, it seems to have been strangely 
dumb regarding the great possibilities of music. Practically 
all of the music of the Greeks was encompassed in two 
octaves or a little more. The central principle of the Greek 
mind, that moderation should be observed in all things and 
that extremes should be avoided at all costs, seems to have 
blinded the Greeks to the possibilities of music in the more 
than seven octaves used by us today, 



The only music the Greeks knew was that used to ac- 
company their poetry, and it was a sickly kind of music, 
Joeing melody only and lacking both rhythm and harmony. 
Tfie only accents in the melody were such as occurred in the 
customary dropping of the voice at the end of a sentence and 
in the observance of the long and short syllables which 
marked the meter of the poetry. There was no harmony to 
enrich the lone melody, for the human race had not yet 
learned to sing or play instruments together, and even the 
simple art of singing a fifth or a fourth apart to make an 
elementary harmony was not learned for more than a thou- 
sand years. 

It must be said to the credit of the Greeks that they took 
an intellectual interest in harmony and developed in a scien- 
tific way the concepts of the harmonic intervals through the 
division of the string monochord, a one-string instrument. 
But this development of the idea of harmony, important as 
it was, got no farther than scientific principles. This is all 
the more surprising, since the celebrated Greek genius ex- 
pressed itself so fully in the other arts of poetry, sculpture, 
architecture and painting. Music seemed to be the one art 
in which they could not excel. 

The first glimmer of harmony is found in the early Middle 
Ages, when some musical genius, greater than the Greek 
musicians, conceived the amazing idea that two voices could 
sing at the same time a fifth apart. As the voice carrying the 
melody sang, another voice a fifth below trailed along, always 
maintaining the same interval. We do not know what 
prompted this reckless thinking, but it would seem to us a 
most natural thing to do, since the ordinary soprano and 
tenor voices have a natural range of about a fifth above the 


contraltos and basses. Such harmony seems to us very ele- 
mental, but when it was first tried more than a thousand 
years ago it must have sounded most unusual and startling. 

Thus began the idea of harmony, but the idea of rhythm 
was still to come. These early churchmen, chanting their 
monotonous music in the churches of the Middle Ages, could 
have learned much about rhythm had they consulted the 
dark tribes of Africa. Here rhythm had already been brought 
to a high state of development, but the religious mind of the 
time regarded rhythm as something sensual and therefore 
to be avoided. 

In the sixteenth century A.D., this type of music reached 
its peak with the celebrated musician of the Church, Pales- 
trina. He did not understand harmony as we do, but he 
developed to its logical perfection the conception that music 
consisted of two or more melodies sung together. He wrote 
music which consisted of a number of melodies sung simul- 
taneously, something like the old madrigal, or the "round/' 
of which the "Three Blind Mice" is the best known today. 
The harmony voices did not simply trail along together, 
a set distance of a fifth or fourth or third apart, but each 
melody set out independently for itself. So ingenious was 
Palestrina that although there were several melodies going 
at once, they were so written that they all harmonized to- 
gether. Such music was like intricate lace, beautifully woven 
together with great cunning and skill. But musicians had not 
yet learned how to write a melody and surround it and 
embellish it with a structure of chords. It was another cen- 
tury before this musical phenomenon was to make its 

As for musical instruments, the musicians had not learned 



to play together. Single instruments had been played for 
thousands of years. Occasionally some of these had played 
an octave apart. But the art of playing together as a group 
was not known until the sixteenth century, and nothing 
approximating the modern orchestra was known until 1600. 
Troubadours, minnesingers and meistersingers twanged their 
guitars and bowed their fiddles, but these performances were 
essentially solo in nature. We even read about miscellaneous 
groups of violins, viols, flutes, oboes, cornettos, fifes and 
drums playing together for Queen Elizabeth of England in 
1561, but we shudder to think what they must have sounded 

Modern music began shortly after the middle of the six- 
teenth century with the budding of opera in Florence, Italy, 
and this was the beginning not only of opera but of the 
orchestra. A group of musicians in this Italian city became 
interested in Greek poetry, and in trying to give their version 
of what it must have sounded like they originated a new 
kind of music known today as opera. They wrote a dramatic 
play in poetry, had it recited to music and then called on the 
musicians of the day to accompany the singing with a miscel- 
laneous group of instruments. The result was not Greek 
drama or Greek music but a new form of musical expression, 
out of which the modern opera and the modern orchestra 
germinated and grew to maturity in the short space of less 
than three centuries. 

These early groups of musicians were not, it is true, much 
like the modern orchestra. They were a miscellaneous lot of 
ancient lutes, lyres, viols and wind instruments, led haltingly 
along by the harpsichord, the grandfather of the piano. They 
did not play as the instruments play today, but produced 

chords as a background for the singer. Their performance 
was much like the performance of the guitar or ukulele today, 
which is used to "chord" to the melody as it is sung. But this 
lowly beginning is not to be made sport of, for it was really 
a remarkable step in music. Here at last was the concept of 
harmony in music, and it was pregnant with amazing possi- 
bilities, as was soon to be revealed. 

At first the music for these performances was not written 
out for each instrument. The group of instrumentalists was 
very unstable, and the early composers had to take what 
musicians they could get. Even if they could have had just 
the instruments they wanted, they were not at all sure what 
instruments they should use. They finally decided that they 
would write a sort of shorthand outline of the music and 
leave the rest to the improvisation of whatever musicians 
happened to be present. This musical outline was called a 
figured bass, and it survived for a hundred and fifty years. 

Another reason for this indefiniteness of the written score 
was the lack of understanding of the individual traits and 
capabilities of the instruments. The musical experience of 
these composers had been restricted to writing for voices, and 
in the beginning they looked upon the instruments simply as 
additional voices. In fact, in many of the musical scores of 
the time, even those which went beyond the figured bass and 
provided separate parts for instruments, the score was 
marked "to be either played or sung." It was the composers* 
idea that any music could be both sung or played on instru- 
ments. When a musician joined the orchestra, he could either 
chord along with the figured bass or take a written part which 
was within the range of his instrument. An early historian 
of music called Praetorius, writing about 1617, says that it 



was customary for the violins and cornettos to play the music 
written in the G clef and for the tenor and bass instruments 
to play the music written in the C, or lower, clef. Beyond this 
there were few directions or customs to follow. 

There is no better way to get an idea of what these early 
"orchestras" were like than to take a look at some of them. 
In 1565 two Italians by the name of Striggio and Corteccia 
collaborated in writing some music to accompany light plays. 
They called for two harpsichords, four violins, three violas, 
six or seven lutes and lyres, about the same number of flutes 
and flageolets, four cornettos and four trombones. Such an 
"orchestra" must have been mostly top and little bottom, 
for there was no bass instrumentation except the lower range 
of the harpsichord. The music, too, must have jumped along 
rather jerkily, for there were so many plucked instruments. 
Even the harpsichord, which gave us the idea for the piano, 
was merely a plucked string instrument, and it was looked 
upon as the most important instrument in the whole group. 

A similar idea was tried out in France by Balthasarini in 
1581. He wrote a ballet and conceived the idea of having a 
little musical accompaniment. For this he wrote a score call- 
ing for oboes, flutes, flageolets, cornettos, trombones, violins, 
viols and lutes. These instruments did not all play at once in 
one ensemble but in small groups, first the strings, then the 
woodwinds, and then the brass, the harpsichord playing 
throughout and furnishing a sort of bass foundation or back- 
ground. That this seemed like a pretty good idea is seen from 
the fact that the famous opera composer Lully exactly one 
hundred years later produced in France the opera "The 
Triumph of Love," and had the strings, woodwinds and 
brass play in separate groups. 


In 1600 two important musical events occurred in Italy 
which influenced music ever after. In this year the first opera 
ever shown in public was performed, and the oratorio was 
given birth. Jacopo Peri produced his opera "Euridice" and 
wrote music for a harpsichord, two lutes, a lyre and three 
flutes. These instruments furnished chord accompaniment 
while the singers sang or recited poetry. Although the score 
called for the instruments named, there were no specific parts 
written for any of the instruments except short passages for 
the three flutes. The only indication that the other instru- 
ments were called for was that the names of the instruments 
were written at the beginning of the score! Peri was more 
considerate than most contemporary composers, for usually 
the composer did no part writing at all, furnishing only the 
shorthand music of the figured bass part and trusting to luck 
that the various instrumentalists could make some music 
from that. 

The first oratorio was given to the world by Emilio del 
Cavaliere and was called "Representation of the Soul and 
Body." This was a sort of religious musical drama, and a 
group of instrumentalists was called in to furnish the chord 
accompaniment. As usual, the harpsichord was one. Others 
were the lyre, lute, two flutes and two violins, an additional 
violin being called for to play in unison with the soprano 
voice throughout. Although these were the principal instru- 
ments Cavaliere called for at the beginning of the score, he 
marked some parts of the music: "To be played by a large 
number of instruments/' This is further evidence of the lack 
of a definite idea of what an orchestra should consist, 

But Cavaliere did set musicians thinking about one impor- 
tant point which was novel for his time. Apparently he was 



not entirely satisfied to have the instruments play monoto- 
nous chords, as they were accustomed to doing. He saw a 
chance to let the instruments enter a little into the spirit of 
the drama being sung. His view was that the instruments 
were not simply there to furnish a chord background for the 
singing they could help interpret the plot of the musical 
drama. Many of the conventional composers of the day 
must have been a little shocked when he said, "Music should 
vary according to the sentiments expressed by the singer." 
Here was attention given to the musical expression of the 
orchestra, an embodying by the orchestra of the ideas ex- 
pressed in the oratorio or opera. This was a new idea which 
had far-reaching effects. 

The first group of musicians which deserves the name of 
orchestra was that called together by Claudio Monteverde 
in 1608, to furnish music for his opera "Orfeo." He wrote * 
for a string section comprised of two violins, two bass viols, 
ten tenor viols, two viol da gambas (cellos), a harp and two 
guitars. The wind instruments consisted of four trombones, 
two cornettos, a small flute, a clarion, three trumpets with 
mutes and three organs. Of course, the harpsichord was there 
nobody seemed to think in those days that an orchestra 
was possible without the harpsichord to lead. This was an 
imposing array of instruments, but they were not handled 
as are the instruments in today's orchestra. The only music 
they had was a figured bass part, and the improvising of such 
a motley group of instruments must have been something 
amazing to listen to. 

But Monteverde learned a lot from this venture, and in 
1624 he composed the opera "Tancredi e Clorinda." In this 
opera for the first time we have a composer writing, not 


simply music to be "sung or played," but music specifically 
adapted to the individual nature of the instrument. He 
amazed both players and audience by having the violinists 
lay down their bows and strike the string with two fingers, 
producing pizzicato. He also had them play a note with 
quickly repeated strokes of the bow, producing tremolo. At 
first the players refused to perform such nonsense, but after 
he had persuaded them to try it they were fascinated with 
the effect, and it is said that when it was played before the 
special audience of princes and other noblemen of Venice 
they broke down and cried. At any rate, here for the first 
time on record is music written for a special musical instru- 
ment, music which is impossible to reproduce vocally. Here 
was the first really instrumental music. Monteverde ex- 
plained that the vocal style was tender, but that he had 
called on the instruments to play the "agitated" style. In 
another place he wrote a descriptive passage which was to be 
played during a combat by the actors, and the music was so 
written that it depicted the strife and fighting. Here was 
"descriptive" music, probably the first of its kind. 

During the next hundred years progress was slow and little 
improvement was made over Monteverde. In fact, most 
composers did not do as well as Monteverde, either because 
they did not know about his advanced ideas or because they 
declined to follow what might have seemed to them a radical 
course. The greatest advance was made in writing for the 
string section. The individualism of these instruments seemed 
to be fairly well understood, and music written for them was 
true string music. Parts were written for first and second 
violin, viola and bass, but to these early composers the cello 
seemed to have no place. 



Although the make-up of the string choir was well estab- 
lished, the instruments which made up the brass choir varied. 
Usually there were trumpets and drums, but the presence 
of the other brass instruments was uncertain. The composers 
were familiar with the trumpets from military use, and since 
the drums always accompanied the trumpets in things mili- 
tary, they thought they should be used together in things 
musical. The horns and trombones were not so familiar to 
them, and they were uncertain about their use. Even when 
they were called into use, their parts were not true brass 
music, but music which any other instrument having the 
same range could have played. 

And as for the woodwinds, they were in an even worse 
position. The flutes were common, both the end-blown flute 
or flageolet and the side-blown flute. Oboes and bassoons were 
also called for but were as often omitted. The clarinet had 
not yet been recognized at all, and the music the woodwinds 
were called upon to play was not individual woodwind music 
at all; it could just as well have been given to the vocalists. 
The harpsichord was looked upon as the most important 
instrument of them all. Composers felt they could not risk 
turning the strings, the brass and the woodwinds loose on 
their own. The harpsichord was needed to hold them all 
together and to cover up their weaknesses. It must be ad- 
mitted that the composers* fears had some justification. Not 
only were the composers of that time ignorant of the re- 
sources of these instruments which would have permitted 
a satisfactory performance without the harpsichord to guide 
and support them, but the instruments themselves were 
crude and imperfect compared to those we have today. As 
the instruments were improved, composers gave them more 


important work to do, and as they experimented with the 
resources of the instruments, they learned to take more 
advantage of what possibilities the instruments possessed. 
The composers and the instrument makers worked side by 
side, each encouraging and inspiring the other. 

Two great musical figures completely dominate the first 
half of the eighteenth century Bach and Handel. The 
instruments they wrote for were fairly well determined and 
settled. They usually had a group of well-disciplined strings, 
and they knew how to write for them and exploit their unique 
resources. Also, they had whipped into shape a stable brass 
section, consisting usually of two trumpets and two French 
horns but no trombones. The side-blown flute had pretty 
well won the battle over the end-blown flageolet, and two 
of these were generally called for. Two oboes and one or two 
bassoons completed the woodwinds. The clarinets and cellos 
were still outside the circle of accepted instruments, along 
with the trombone. The use of the harpsichord was still a 
crutch which kept these two great composers from venturing 
too boldly in their writing. And lutes and lyres and other 
quaint instruments were still admitted into respected ranks. 

But, although these three choirs strings, brass and wood- 
winds were becoming stable in the number and kinds of 
instruments, and although the instruments themselves were 
written for with due attention given to their individuality, 
the treatment of the whole was rather different from the 
harmonic and symphonic writing which developed later. 
Bach stuck to the polyphonic style of Palestrina and created 
some great masterpieces in this kind of writing, Handel's 
music was more like the music we are familiar with today in 
its harmonic structure and the distribution of the notes of 



the chord to the various choirs and instruments. Both 
orchestras had attained solidity and balance but lacked the 
variety and contrast which later composers used. Bach and 
Handel had not yet learned the great tonal spectrum of their 
instruments and used a limited palette. They took more of an 
intellectual view of their music and worked it out meticu- 
lously in form and pattern, especially Bach, but neglected 
to enrich it with color, contrast and variety of effects. 

Haydn is often called the father of the symphony because 
he weeded out the undesirable instruments and left us with 
an instrumentation which, except for a few changes, is the 
orchestra we use today. Gluck, a famous writer of opera, 
certainly pointed the way, as opera writers have done since 
the birth of opera in Florence, for he admitted the ostracized 
trombone, sometimes used piccolo, used the clarinets 
although he treated them as if they had been oboes refused 
to lean on the harpsichord for support, and filled out the 
percussion section with bass drum, side drum, cymbal and 
triangle. It cannot too often be emphasized that not only 
did the first orchestra enter music through the back door of 
opera, but as each instrument or instrument effect was 
added, it usually had to come in the same way. 

The writer of opera is a showman as well as a musician. 
He strives for new effects. When Monteverde taught the 
string instruments to play pizzicato and tremolo, he was 
striving for an effect which would heighten the action of the 
drama. This search for the novel has always characterized the 
successful opera writer. Naturally, he was always on the 
alert for new instruments which could produce new colorings 
and effects. As the plots for operas had to be built up to 
climaxes, so also did the music. Opera audiences have been 


taught to want strong contrasts and surprising effects. The 
early composers of opera saw that the music could assist 
in the total effect. They studied to make the instruments they 
had do things they had never done before, and they sought 
out new instruments which could do things the other instru- 
ments could not do. The opera orchestra, therefore, was a 
sort of training school for the symphony orchestra. No 
instrument could hope to make its debut in the symphony 
if it had not previously served an apprenticeship in the opera. 

Haydn followed Gluck in his use of instruments in the 
opera but stuck pretty closely to a more conservative in- 
strumentation for the symphony. He used clarinets in opera 
but not in the symphony. He and Mozart both used trom- 
bones in their opera but refused to give them a place in the 
symphony. Mozart was willing to try out the basset horn, 
or alto clarinet, in his "Magic Flute" but did not think it 
suitable for the symphony. Four horns were used in opera, 
but in the symphony two were standard. The symphony 
orchestra of Haydn and Mozart is the so-called classical 
orchestra consisting of string quartet, two trumpets, two 
horns, two flutes, two oboes and two bassoons. It lasted until 
Beethoven, but he let down the bars and included a number 
of instruments which he thought had served their apprentice- 
ship and which he needed to give musical expression to his 
more romantic thought. 

When Haydn turned over the orchestra to Beethoven he 
had done a great deal more than catalog the instruments 
which should be admitted to the symphonic ranks. He also 
developed to a high point the classical method of orchestrat- 
ing. To the solidity and balance of the orchestras of Bach 
and Handel, he added some contrast and variety. He spread 



his chords over a greater number of instruments and built the 
chords in a variety of ways, integrating and solidifying the 
harmonic structure. To his pupil Mozart, though, must go 
the credit for first using a specific tone coloring. It was this 
appreciation of tone coloring which prompted Mozart to use 
clarinets, for he saw that they contributed a contrasting 
coloring which was new and different from anything in the 
orchestra up to this time. He taught his teacher Haydn the 
use of clarinets, and through this schooling from his pupil, 
Haydn learned in his old age to use, not only the clarinets, 
but also other woodwinds. He demonstrated how to write 
individual and characteristic music for the woodwind choir. 
The passing of the harpsichord was a great boon to the 
orchestra. How it managed to hang on for two hundred years 
is difficult for us to understand. Many of the composers were 
players of the harpsichord, and they found it convenient to 
sit at the instrument and direct their compositions by play- 
ing along with the orchestra. If a player faltered, the com- 
poser could play his part on the harpsichord and set him 
on his feet again. It was also used as a rhythm instrument, 
and the composer could beat the tempo by playing chords 
in measured -accents on the harpsichord. But it had to go 
when music developed beyond the polyphonic style and took 
on contrast and color, for this instrument lacked these 
qualities. It was not playable both soft and loud, as is the 
piano, for notes could not be struck with varying degrees of 
loudness. The strings were plucked by quills when the keys 
were struck, and every note had the same monotonous vol- 
ume. Chords could not be sustained, and the only way the 
breaks in the musical pattern could be filled in was with 
cadenzas and other figured embellishments. It lent itself 

therefore to polyphonic music, in which a continual inter- 
weaving of melodies occurred. When great and daring har- 
monic structures were built up in the music, the harpsichord 
floundered and gradually became more and more outmoded 
until it passed out, and with it went the old figured bass. 
A new day in music dawned with their passing. 

The first two symphonies of Beethoven were hardly to be 
distinguished from those of Haydn, but in the third, the 
"Eroica" of 1805, Beethoven gave expression to a new kind 
of music. The old classical music had been largely intellectual, 
and although it was beautiful in its own way, the great 
Romantic movement of Europe sought expression in music 
and called for new instrumentation and orchestration. The 
Romanticist was not satisfied with intellectual pleasures but 
craved something which stirred the feelings. In the symphony 
Beethoven satisfied this urge by mixing in a greater wealth 
of tonal coloring. He welcomed the trombone, the cello, 
the piccolo, and even the deep-toned contrabassoon, for 
they contributed a more colorful tonal palette. He wrote 
in a different style, also. He followed the operatic writer by 
building up great contrasts and climaxes and introducing 
a greater variety of treatment. 

"Beethoven was the first to add the third and the fourth 
horn in the symphony, and he taught the instruments of the 
classical orchestra to play music for which they were not 
thought to be capable. He wrote solos for the string basses 
which the bass players at first refused to attempt, saying 
such music was impossible. He sought out new resources in 
all the instruments and mixed his tonal colors in novel ways. 
By the time he had completed his great Ninth Symphony 
he had brought the symphony orchestra to a high level of 



performance which in many ways has scarcely been sur- 
passed, if equalled. In solidity, balance, contrast and variety 
of effects he wrote the final chapter, but in tonal coloring 
there were masters to come who excelled him. 

Other great emotional composers followed Beethoven who 
enriched symphonic music with their musical ideas and ex- 
quisite workmanship. Among these are Schubert, Schumann, 
Mendelssohn, and that wizard in orchestration and tonal 
color, Berlioz. Although Berlioz himself played no instrument 
except the guitar, he understood the instruments of the 
orchestra better than the musicians themselves. He was the 
first composer who wrote the low pedal notes for trombones. 
He knew trombonists were unfamiliar with these notes and 
wrote on the manuscript, " These notes are on the trombone 
and the players must get them out." He was also the first to 
use violin harmonics in full harmony. To achieve a peculiar 
muffled clarinet effect, he once specified that the instrument 
be put in a leather bag and played. He was constantly striv- 
ing for new combinations of instruments and often went to 
excess in requiring an inordinate number of certain ones. 
But if he did not always achieve a musical effect, he un- 
erringly achieved the effect he had in mind, for scarcely any 
composer, before or since, understood so well the individual 
resources of each instrument and the unique tonal color and 
effects of which each was capable. His experiments are a 
storehouse of experience for composers who followed him, 
and many have profited by his pioneering. 

But sharing the honors with these great masters were the 
instrument makers. In 1832 Boehm brought out his new key 
mechanism for the flute, which also became the basic system 
for clarinets and contributed improvements to oboes. It is 


obvious that composers and musicians could do much more 
with twenty-three keys than they could do with eight. 
Boehm's amazing key mechanism made woodwinds chro- 
matic in all keys and made easy passages which before were 
played with the utmost difficulty; and it made possible many 
passages which were totally impossible on the old woodwind 
mechanism with eight keys. 

Piston valves were invented in 1815 and rotary valves in 
1827. These inventions bridged the wide gaps in the open 
scale of the simple trumpets and horns and added a great 
wealth of playing possibilities which the older composers 
could not entertain. New instruments were now perfected 
which formerly had lain in disuse because of their limitations, 
and the tonal coloring was made still more various and 
brilliant. The English horn was introduced by Rossini in 
"William Tell" in 1829, and the bass clarinet was included 
by Meyerbeer in "Les Huguenots" in 1836. 

And with the invention by Sax in 1842 of the family of 
saxhorns, the brass choir for the first time was given a good 
bass voice in the brass bass tuba. Composers before Wagner 
had been handicapped by having too few bass voices. For 
many years the only wind basses were the bassoons and an 
occasional bass trombone. The old serpent and ophicleide 
bellowed and brayed for two centuries, but they were in- 
accurate in intonation and of poor musical quality. The great 
success of Wagner in writing for the brass instruments was 
partly due to the addition of the bass tuba. This rich sonorous 
voice, added to the new bass clarinet, augmented the bottom 
of the orchestra and gave it great depth and beauty. 

Erard's improvement of the harp and Sax's invention of 
the family of saxophones further enriched the tonal coloring 



and gave variety which has enhanced the orchestra. Bells, 
the xylophone, marimba and other percussion instruments 
were also used to good advantage in creating musical effects 
unknown even to Beethoven. 

Wagner brought the orchestra to a new peak in musical 
expression through his consummate mastery of instrumental 
technique and uncanny feeling for color and effect. Weber 
and Meyerbeer, two great opera composers, blazed the trail. 
They were great dramatists and explored the orchestra for 
dramatic effects. Wagner is especially indebted to Weber, 
for the latter pointed out a musical realm which Wagner 
exploited to the utmost. Wagner's skill in handling brass 
had never been equaled before, and he did many daring 
things in dividing the strings into many separate parts and 
creating other beautiful effects which were new in the 

His harmonic structure was the most complicated and 
intricate that had ever been constructed. In his "Ring" 
series he scattered notes over the entire seven octaves, dis- 
tributing them judiciously to the various choirs, to be 
sounded together in one great, overwhelming diapason of 
sound. Fifteen notes are given to the woodwinds, which the 
flutes, oboes, clarinets and bassoons sound together. To the 
trumpets, horns, trombones and tubas of the brass choir he 
gives seventeen notes, making a total of thirty-two notes 
in the harmony played by the woodwinds and brass alone. 
Add to this the multiple division of the strings, equal to that 
of the brass and woodwinds, and we have an absolutely over- 
powering effect of wonderful beauty. 

After Wagner it might be thought that there remained 
nothing further to try. But, although composer, musician and 


instrument maker had traveled a long way since Monte- 
verde, Bach and Handel, new adventures in music have 
been made recently. This advance has been made in develop- 
ing exquisite tone color. Ever since Liszt, Chopin and the 
other tone poets, composers have been absorbed in creating 
new color schemes. Their maxim was beauty for beauty's 
sake, and they have created some amazing compositions. 
Of these composers, Debussy and Richard Strauss are typ- 
ical. They go in for tone poems and descriptive music. The 
perfection of musical instruments in the twentieth century 
and the use of newer instruments such as the saxophones 
have given them greater musical resources for carrying out 
their ideas. They have also called in some unusual instru- 
ments for their specific purposes, Strauss, for instance, has 
revived the heckelphone, the baritone voice of the oboe 
family; he has also written for alto and bass flutes, Efc and 
D cornets, alto, tenor and baritone tubas, and even the 
sarrusophone, a sort of brass bassoon. 

Such, in broad outline, has been the development of the 
orchestra. Much of its course has been determined by acci- 
dent, and speculative "ifs" mark every step. If the Italians 
of Florence had not experimented with Greek drama com- 
bined with music, we might never have had anything like 
the orchestra, and we might still be in the era of the trouba- 
dours and minnesingers. If the writers of opera had not been 
such prospectors for the gold of musical effects, we might still 
be twanging lutes and lyres and blowing our lungs out on 
serpents and ophicleides. If the romantic urge had not taken 
hold of music, we might still be spinning out the intellectual 
intricacies of Bach, Handel and Haydn. If inventive genius 
of the instrument makers had not produced the ring key of 



the Boehm system flute, the piston and rotary valves of 
Blumel to say nothing of needle springs and hinges mounted 
on posts we might still be playing the music of Beethoven 
and his predecessors and be unaware of the beauties of 
Wagner, Debussy and Strauss. 

The gripping story of how each instrument worked out its 
own destiny and won a place for itself in the great symphony 
orchestra and concert band of today is told in the following 



The Fiolin Family 


Now THAT THE VIOLIN has "made good " in the symphony 
orchestra, many countries and peoples claim it as a 
"native son." Although the true ancestry of the violin will 
probably never be determined to the satisfaction of all, there 
seem to be about three chief contenders : Arabia, Wales and 

Little did the Moors realize, when they invaded Spain in 
711 A.D., that they would precipitate a fight about the 
ancestry of the violin that would reverberate down the 
centuries of history for over a thousand years. One of the 
musical instruments which these fierce fighters brought with 
them was the Arabian rebab, a small stringed instrument 
played by plucking the strings. Shortly after arriving in 
Europe the rebab became a favorite of the troubadours of 
Spain, France and Italy. This is the famous troubadour 
fiddle, celebrated in story and song of the Middle Ages, 

Unquestionably the rebab was a plucked instrument when 
it arrived in Europe, for the ancient peoples of the Eastern 
world never did develop the art of sounding an instrument 



by bowing. This is a fact, in spite of the widely accepted fable 
that Nero fiddled while Rome burned. Nero may have plucked 
a lyre, but he never bowed a fiddle. It wasn't long after the 
rebab was taken up by the troubadours, however, until cer- 
tain innovators began playing it with a bow, and it is gener- 
ally conceded that bowing the fiddle became common toward 
the latter part of the eighth century. 

Venantius Fortunatus, the poet, started a lot of argument 
when he wrote, back in 617 A.D., an ode in which he referred 
to the "chrotta Brittanna." Many think he was referring 
to the Welsh crwth, a sort of primitive lute. No doubt this 
instrument was at first plucked, but as far back as we are 
able definitely to trace its history it was played with a bow. 
Therefore many look upon the crwth as the father of the 
modern violin, and others who will not grant the complete 
title admit that the crwth is the first instrument ever played 
with a bow. Even this honor should be enough to make the 
Welsh crwth immortal. 

There is another school which will have neither of these 
instruments. These critics point out that while we find the 
Welsh crwth and the Arabian rebab both played with a bow 
in the eighth century, neither has a sound chest like that of 
the violin, and that this feature and not bowing is the dis- 
tinguishing mark of the violin. They go on to say that the 
violin has ribs between the back and sound board, as dis- 
tinguished from the lute, which has a sound board glued 
directly to an arched back without the intervening ribs. 
Both the rebab and the crwth are lutes, they say, and regard- 
less of when they were first bowed, neither of them can 
qualify as the ancestor of the violin, for bowing has nothing 
to do with earmarking the violin lineage. 


Judged on this basis, the Greek kithara is the first violin, 
for it is the first instrument we know of which used a sound 
chest constructed with sound board and back separated by 
ribs. It is admitted that this instrument was not bowed by 
the ancient Greeks, but after all this point is nonessential. 
The important fact is that in the Greek kithara we find the 
first sound chest of the violin construction. Besides, there is 
philological evidence. It is easy to see the resemblance be- 
tween the words kithara and guitar, the latter being a name 
often used by the troubadours in referring to their fiddles. 
All of this speculation is very interesting, but those who 
require more solid ground for a take-off on the histofyof the 
violin will prefer to start with the troubadour fiddle. From 
this point there is only one hurdle before the violin is reached, 
and this is the viol family. The viols descended directly from 
the troubadour fiddle, and the violins in turn descended from 
the viols. The latter two families are quite similar, differing 
principally in the fact that the violins do not have such deep 
ribs, have swelled or curved backs instead of flat, and have 
a much better tone. One characteristic of the viols has sur- 
vived to this day in the bass of the violin family, for some 
string basses are still found with flat instead of swelled backs. 
The first true violin was made by Caspar Tieffenbriicker, 
born of German parents in the Alps mountains of Italian 
Tyrol about 1467, or twenty-five years before Columbus 
discovered America. His instruments are said to have been 
heavily inlaid and ornately decorated and must have excelled 
in appearance more than in musical quality. By 1520 a 
number of violin makers, building on the work of Tieffen- 
briicker, had gravitated to Brescia, a town in Lombardy less 
than fifty miles from the Tyrol border, and had founded 



what was soon to be known as the Brescian School. This 
group of craftsmen flourished for one hundred years and 
boasted such men as Gasparo da Salo, Maggini, Kerlino, 
Zanetto, Cortesi and Peregrino. 

The finest of all violins were produced about forty miles 
from Brescia, at Cremona. This little town on the banks of 
the Po River began attracting attention because of its fine 
violins about 1550. Illustrious names were those of Andrea 
Amati, his son Geronimo, and Geronimo's son, Niccolo, 
three generations of great craftsmen who brought the Amati 
violins to such a high stage of development. One of Niccolo's 
most talented pupils was a young man named Antonio 
Stradivarius, born in the midst of the violin-making industry 
of Cremona about 1645. He studied under Niccolo until he 
was about twenty years old, when he left the Amati tutorship 
and began making violins after his own ideas. For the next 
fifteen years, however, his violins still strongly resembled the 
Amati models. Gradually he progressed to the so-called 
"Long Strads," and by 1700 he had pretty definitely formu- 
lated the ideas and methods which resulted in the famous 
Stradivarius violins. The violins made between 1700 and 
1730 were his greatest; from 1730 until he died in 1737 he 
made few violins, and these were of varying quality. Itis no 
wonder, for by 1730 Stradivarius was about eighty-five years 
old, and a number of the violins which he commenced were 
finished by his son and by his pupils. 

Other great names associated with the Cremona School, 
besides the Amati and the Stradivarius families, were the 
Guadagnini, Ruggeri, Guarnerius, Bergonzi and Storioni 
families. Carlo Bergonzi was a famous pupil of Antonio 
Stradivarius who undoubtedly finished some of the late 

Stradivarius violins started by the master. Of the Guarnerius 
family, Joseph was the most famous, having made a fine 
violin used after his death by the greatest of all violin virtu- 
osos, Niccolo Paganini. The middle of the eighteenth century 
had hardly passed when the glory of the Cremonese School 
began to fade, marking two hundred years of great achieve- 
ment, the golden age of the violin. No other one group of 
violin craftsmen has ever equaled the famous makers of 
Cremona, but it is not true that the art of violin making died 
with them, for there have been many wonderful instruments 
made since then, and they are still being made today, 

Many of these old violins are now worth thousands of 
dollars. Some of them have been sold for as high as $50,000 . 
and others have brought $15,000, $25,000 or $40,000. But 
just because an old violin has lain in the attic for generations, 
or has been handed down from father to son, is no sign it is a 
valuable Stradivarius, Guarnerius or Guadagnini. There are 
many "phony" Stradivarius violins circulating around, and 
it takes an expert to tell whether they are genuine or not. If 
all the so-called Stradivarius violins were gathered together 
their number would be so great that it would have required a 
score of Antonio Stradivariuses to make them, even though 
they all lived to be ninety, as did the great master. Experts 
say that Stradivarius made about 950 violins during his whole 
lifetime, besides about one hundred and fifty violas and 
cellos. There are today authentic records of about 425 
genuine Stradivarius violins, besides seven or eight violas. 
There no doubt are others still in existence, but the chance of 
finding one is slight. 

Labels mean little to the experts; their trained eyes and 
ears look for more dependable information, such as tone 



(Above)'. Viola da braccia viol held against shoulder from which 
the violin descended. Note short, heavy bow. (Below): Viola da 
gamba viol held between the knees from which cello descended. 
Note heavy, clumsy bow. 


Although all violins look alike to most people, there are individual 
differences in size and shape of upper and lower bouts (eonvexly 
curved parts of body), the shape, size and location of the / holes, 
length and width of entire body, and other features readily dis- 
cernible to the expert. 


quality, varnish, workmanship, and many little telltale clues 
which only the experts see at all. There are many copies of 
famous violins made copies made to the smallest details. 
The shape and position of the ff holes, the size and shape of 
the upper, middle and lower bouts and their size relative to 
each other, the length of body from base of button to tail 
pin and length of button to notch of ff holes, the depth of 
ribs at all points, the curve of back and top all these 
measurements have been meticulously duplicated to small 
fractions of an inch, and yet violin authorities can tell they 
are only copies and not originals. The only safe way to avoid 
swindles in buying genuine old-master violins is to consult a 
reliable authority. 

There is a great deal of hokum surrounding the violin, 
especially the violin of the great masters. It is popularly be- 
lieved that what made the old violins good was a collection 
of trade secrets which died when these great old craftsmen 
died, just as the art of making stained glass died with the 
thirteenth-century Gothic cathedral builders. There is much 
talk about varnish, as if it were some sort of magic bath 
which could transform an ordinary violin into a great master- 
piece. Since no one today is able to make a varnish which 
will perform this magic trick, several ingenious stories have 
arisen which lay the blame for our alleged inability to make 
violins equal to those of the old masters, to our lack of the 
proper varnish. One story has it that the resin used by the 
old masters to make their magic varnish came from a certain 
balsam fir tree of northern Italy. Owing to the great demand 
for the resin from this tree, it was tapped excessively and the 
species finally became extinct. For about two hundred years 
this tree has ceased to exist; for about two hundred years 


there have been no violins made which equal those of the 
masters: therefore it is evident that it was the varnish which 
was the secret of great violin making! 

Another easy explanation is the method of curing the wood, 
It is claimed that the process of drying and seasoning the 
wood was the secret of the old masters and that we do not 
know today how this was accomplished. Closely allied with 
this story is the fable about the age of the finished violin. 
Certain magic has been attached to age, as if an ordinary 
violin could be made a masterpiece if only it could attain a 
ripe old age. If this were true, any old cigar-box violin or 
cornstalk fiddle could hope to be a tolerably good musical 
instrument after two or three centuries. Much speculating 
has also been done about the shape of the violin and its 
various parts, particularly the curvature of the back and top, 
the shape and position of the ff holes, the relative sizes of the 
upper and lower bouts, and so on. 

Most authorities on the violin now agree that the achieve- 
ments of the old masters were due, not to any magic formulae 
or trade secrets, but to more prosaic reasons. These craftsmen 
were born into the violin-making art. They grew up with it, 
They thought of nothing else. To be a great violin maker 
was the ambition of every boy in the village. This singleness 
of purpose is at the basis of all great achievements in art 
and craftsmanship. Apprentices were given long and exacting 
training. Journeymen labored years before they were gradu- 
ated into masters of the craft. All their work was done with 
infinite care, unflagging patience, consummate skill and an 
inspired desire to achieve great things, Among these crafts- 
men there grew up a distinct esprit de corps similar to the 
religious fervor which was the driving force behind the con- 



ception and erection of the great cathedrals in the thirteenth 
century, and similar to the artistic standards and devotion 
to a craft which characterize the wood-carving Passion 
Players of Oberammergau. There was just one end in life 
for the Cremonese workman, and that was to make each 
violin a masterpiece and to make each successive instrument 
a bit better than the former one. 

That these old masters reached a high stage in the develop- 
ment of this instrument is attested by the fact that their 
violins never have been surpassed, if ever quite equaled. 
Many experiments have been made during the past two 
hundred years without contributing a single important im- 
provement. Violins have been made of all kinds of wood, 
seasoned and treated in all kinds of ways. Many materials 
other than wood have been used, but without success. A 
modern invention is the aluminum violin, but it is generally 
conceded that this type of instrument reproduces too many 
of the higher harmonics and is of too brilliant tone. The 
shape has been varied in many ways, and the dimensions 
have been changed, but it seems that the old masters must 
have been over all this ground before, for always these in- 
novations are found to be no better than and usually not as 
good as the standards set up over two centuries ago. 

With slight variations, the best violins made today are 
duplicates of those made by the masters. Seventy parts are 
used, the most important being the straight-grained spruce 
top and the curly maple back. The spruce top is always of 
two pieces, sawed on the quarter from the same log, so the 
grain matches on each side of the center where the two pieces 
are glued together. The back is usually made of two pieces 
in the same manner, although some fine violins are of one- 


piece back. The proper thickness and curvature of the top 
and back require shrewd judgment, and it is here that the 
masters excelled. Wood varies in density, resiliency and grain 
structure, and these old makers attained their marvelous 
results by patient shaving of the wood here and there and 
by indefatigable trial and error in testing results. After the 
top and back are shaped, the / holes are cut in the top, an 
operation of great importance. The maple sides are then 
shaped by a heating process. The linings and six inside blocks 
are cut from spruce, and the fingerboard, pegs, tailpiece and 
saddle are fashioned from ebony. All parts are fitted together 
with fine accuracy and glued. A light oil varnish is used for 
finishing, heavy varnishes having been found to destroy the 
mellowness of tone desired. 

Contemporary with Antonio Stradivarius were such great 
violinists as Corelli and Tartini; but Stradivarius was dead 
nearly fifty years when the greatest of all violinists was born 
Niccolo Paganini. Although unquestionably endowed with 
amazing ability, Paganini could not have achieved his great 
technique had it not been for Francois Tourte, the great 
French "Stradivarius of the bow." The masters of Cremona 
had brought the violin to a high stage of development, but 
they seemed to have neglected the bow. No doubt Corelli and 
Tartini would have developed greater technique and would 
have composed more brilliant solos and concertos for the 
violin could they have availed themselves of the improve- 
ments in the violin bow made by Tourte after they were 

Ten years after Stradivarius laid down his tools and glue- 
pot and was buried in Cremona with fitting honors, Francois 
Tourte was born in Paris. He early became interested in 



improving the violin bow. His first specimens were made 
from staves of sugar casks, but he experimented with many 
kinds of wood and finally decided Pernambuco wood was 
best suited to his requirements. How well founded this 
conclusion was is attested by the fact that no other material 
has been found which quite equals this red, rare wood from 
Brazil. Before he was thirty, Tourte was famous as a maker 
of fine violin bows. The bow before Tourte was short, heavy 
and clumsy; he made it light, flexible and resilient. He worked 
out a delicate balance and graceful shape and fixed its length 
at a fraction over twenty-nine inches. He also invented the 
movable nut for loosening and tightening the hair. The hair 
itself was carefully selected and placed in the bow, and a few 
less than 150 hairs were found to give the best results. 

Violins at the beginning of the seventeenth century were 
of two types: the viola da gamba, played by holding between 
the knees, and the viola da braccia, played by holding against 
the shoulder. Of each of these two types there were several 
sizes, but during the process of development a number of 
these dropped out and the well-known members of today's 
violin family survived. The treble viola da braccia became 
the violin, while the alto viola da braccia became the viola. 
The tenor viola da gamba survived the selective process and 
became the violoncello, or cello. Our present-day double-bass 
has had quite a struggle for its place in the orchestra, having 
been almost nosed out by the ancient bass viol, with the 
deep ribs and flat back. This obsolete instrument survived 
until well into the eighteenth century, and some of its 
characteristics are still seen in a certain type of flat-back 
double-bass found in the modern orchestra occasionally. 

Members of the violin family had been used for centuries 


by the troubadours and minnesingers of Europe as accom- 
paniment for their songs, but the first record we have of the 
use of the violin in the orchestra is in 1565, when Striggio 
and Corteccia scored for a group of instruments used to play 
between the acts of early Italian opera and to accompany 
light plays and ballets. In this odd assortment of instruments 
we find four violins and three violas, besides six lyres and 
lutes and some flutes, cornets, trombones and other miscella- 
neous instruments of the day. In France sixteen years later 
we find Balthasarini scoring for a primitive-type orchestra 
used to play accompaniment for his " Ballet Comique de la 
Reine," first performed in Paris in 1581. In this collection 
of instruments, which hardly deserves the name of orchestra, 
he used ten violins, a viola da gamba, and some lutes, besides 
an assortment of woodwinds and brass. 

It was the custom in Florence during the notorious reign 
of the house of Medici to stage big celebrations when a 
member of the reigning house was wed. Italian opera during 
this period was having growing pains, and some kind of 
operatic performance was usually a part of the festivities. 
In 1600 when Marie de' Medici was married to Henry IV, 
King of France and Navarre, Peri and Rinuccini staged the 
celebrated "Euridice," in honor of the event. This is the 
first opera of which we have the entire score, and the instru- 
ments used in the orchestra are an interesting lot. The only 
member of the violin family found in this so-called orchestra 
was the viola da gamba, but there were a number of near- 
violins, such as the lute, the theorbo (large lute) and the 
lyre. Add to these a harpsichord and three flutes and try to 
imagine such an orchestra playing chords together as a 
sort of musical background to dramatic recital of how Or- 



pheus tried in vain to waft his dead wife Euridice back 
from Hades on the wings of song, and you get a rough idea 
of the orchestra of the time and the kind of company the 
violin kept in those days. 

The violin family had a little more prominent part in the 
birth of the oratorio. In 1600 Emilio del Cavaliere produced 
in Rome his celebrated first oratorio, bearing the prodigious 
name of "La Rappresentazione dell' Anima e Corpo." The 
violin used was called upon by Cavaliere to play in unison 
with the soprano voice. The viola da gamba, the only other 
member of the violin family found in this famous cast, joined 
in with a double-lyre, a harpsichord, a bass lute and a couple 
of flutes. 

The term orchestra is used very loosely when we apply it 
to such motley groups of instruments as the foregoing ex- 
amples. In 1608, however, Monteverde, an Italian opera 
composer, wrote an orchestra score for his opera "Orfeo," 
which called into service a rather presentable group of in- 
struments. Among them were "two little French violins," 
two viola da gambas, ten tenor viols, two bass viols, besides 
two large guitars and a number of brass and woodwind in- 
struments. This group is really the first which deserves the 
name of orchestra, and to Monteverde is usually given the 
honor of having founded the orchestra. 

Monteverde seemed to have a particularly good under- 
standing of the possibilities of the violin family, and in 1624, 
in his opera "Tancredi e Clorinda," performed in Venice 
before the Italian nobility, he made the violins do tricks 
which overwhelmed his audience. It was at this time that he 
introduced pizzicato and tremolo effects. 

Following Monteverde, composers and conductors grad- 


ually learned more about the use of the violin family. The 
story of this development is an interesting one of trial and 
error. Many experiments were tried and abandoned, but 
others became standard practice from then until today. 
As was true of all instruments, the violins were looked upon 
in the beginning as voices, to be used simply for singing, 
just as human voices. Gradually, however, the violins, along 
with the other instruments, were regarded as new vehicles 
of musical thought and were boldly experimented with in 
order to exploit their full possibilities. They became the 
famous "string quartet," consisting of first violin, second 
violin, viola and contrabass. With the recognition of the 
place of the cello in this string choir, the violin family finally 
became the famous " string quintet" of the orchestra, con- 
sisting of the soprano first violin, the mezzo-soprano second 
violin, the alto viola, the tenor cello and the bass contrabass. 
Two important composers of Naples contributed much to 
this development. Scarlatti at first had cellos and violas play 
in unison with the bass, but occasionally wrote an independ- 
ent part for the viola. He soon saw the beauty of distribut- 
ing parts to various members of the string choir, and he 
began writing parts for violin, viola and bass. That this style 
of writing is highly satisfactory is shown by the fact that it is 
used frequently by all composers since Scarlatti. Later he 
diversified his writing still more by distributing four notes 
of a chord to the strings, giving the treble part to the first 
violin, the alto part to the second violin, the tenor part to 
the viola, and the bass part to the bass. Stradella, a fellow 
countryman and contemporary of Scarlatti, often wrote 
parts for two solo violins and a solo cello, these lead voices 
being accompanied by violins, violas and basses which filled 



in the harmonies. This style of writing appealed to Bach 
and Handel and was used in their compositions known as 
the "concerti grossi." 

About this same time the great French opera composer 
Lully was using as the foundation of his orchestra a body of 
strings. Violins played the upper parts, while violas played 
the middle and lower parts, trumpets, flutes, and oboes 
being used for creating greater volume of tone. In Venice 
about this time Legrenzi wrote for a rather peculiar orchestra 
consisting of nineteen violins, two violas, two viola da gam- 
fa as, four large lutes, two cornets, three trombones and a 
bassoon. The lack of string basses and of woodwinds was a 
serious defect of this instrumentation, and this type of 
scoring gradually died out after Legrenzi. >Cr 

Two great musical figures completely dominate the first 
half of the eighteenth century: Johann Sebastian Bach and 
Georg Friedrich Handel. Bach was most at home on the 
organ, and the bulk of his compositions was for this instru- 
ment, but he cannot be accused of neglecting any of the 
instruments of the day. He was a most inquisitive man, 
and he tried about everything which offered any chance of 
producing music. We find him writing for violins, violas, 
cellos, double-basses, violino piccolo (a small violin tuned 
a minor third above the violin), viola d'amore (a tenor viol 
with seven gut and seven steel strings vibrating together), 
viola da gamba and lute, besides a flock of woodwinds and 
brasses. Bach apparently didn't think even these instruments 
of the day were all that were needed, for he was the inventor 
of the violoncello piccolo, a small cello. 

Bach's use of the string instruments follows the practice 
of his predecessors. His first "Brandenburg Concerto," com- 

posed in 1721, follows the practice established by Scarlatti 
of writing the score around the string quartet, this main 
body of strings being aided by the double-bass and the 
violino piccolo, besides horns, oboes and harpsichord. Hfe 
also followed the lead of Stradella by featuring several solo 
instruments and supporting them with a full harmony choir 
of strings and other instruments. That he knew the resources 
of the violin as a solo instrument is shown in a number of 
compositions, notably in his "Chaconne," and more popu- 
larly in his "Air for G String," originally written as a move- 
ment in the Suite No, 3 in D Major. Contemporaries of 
Bach were Corelli and Tartini, and while it is not known 
how well acquainted he was with these great violin virtuosos, 
we do know he was a warm admirer of Vivaldi. No composer 
could have written that tremendous violin composition 
"Chaconne" if he had not been thoroughly and intimately 
acquainted with the resources and potentialities of the 

Handel often adopts the ordinary arrangement of first 
and second violins, violas and basses, using oboes and 
bassoons to contrast with the strings, the oboes doubling 
violin parts, the bassoons playing with the basses. In his 
"concerti grossi" he followed the practice of Stradella, or 
rather of Corelli, who perfected and made classical this kind 
of scoring for the orchestra and from whom Handel is sup- 
posed to have learned it during his sojourn in Rome in 1708. 
Handel is credited by some with being the first to bring out 
the unique beauties of the cello, and he did many bold and 
effective things with the viola. 

Besides learning how to distribute the notes of various 
chords among the capable members of the violin family, 



composers and conductors learned new refinements in the 
actual playing and handling of the strings. Johann Karl 
Stamitz, the Bohemian violinist and conductor, became 
director of the famous orchestra for the Elector of Mannheim 
in 1745, and he devoted himself to the task of drilling and 
disciplining the string players to a point of perfection un- 
known before his time. He taught the strings to bow in 
unison and to play with precision and refinement. He drilled 
them in phrasing and in playing as a great body of strings, 
either piano or forte. He showed his contemporaries how to 
produce a quality of tone from many violins which sounded 
nothing like an equal number of violins playing together in 
the ordinary way: he produced a distinct ensemble tone 
which was a revelation. His orchestra became the best in 
all of Europe. The famous composers and conductors of the 
day, such as Mozart, listened to this great body of strings 
and went away amazed and deeply influenced. 

When one calls the roll of the instruments of the orchestra 
before Haydn, one is somewhat surprised to find so many 
queer and unusual names. After Haydn, though, we are 
always among old friends. We no longer encounter lutes, 
lyres, theorbos, viola da gambas, viola d'amores, violino 
piccolos and cembalos. It was Haydn, as has been stated, 
who with almost uncanny insight selected the instruments 
worthy of inclusion in the orchestra and threw the others 
into the discard, from which not a single one has been able 
to stage a comeback into the respected ranks of the orchestral 
instruments. In his first symphony in 1759 he does not 
include the cello among the strings, but in his famous Sym- 
phony in D, written in 1795, we find the cello in its rightful 


The cello's place was not secure until some time after 
Haydn, however. Even Beethoven did not always use it, 
for it is missing from his First Symphony, written in 1803. 
Two years later, though, we find Beethoven has added the 
cello in his great "Eroica" symphony. The contrabass was 
first brought into prominence by Beethoven in his Fourth 
Symphony, at the end of the last movement. This passage 
contains some of the most difficult playing ever written 
for the string bass. At first it provoked much criticism as 
being beyond the resources of the instrument, but Beethoven 
as usual had satisfied himself about what the contrabass was 
capable of. Other composers followed his lead, and before 
long such scoring was taken for granted and was played by 
the musicians without protest. 

Other composers than Monteverde have contributed addi- 
tional effects on the violin. Gluck in his "Armide" was the 
first composer to call for the use of the violin mute. Weber 
enriched the tapestry of violin playing by dividing the first 
violins so part of them played one strain while the rest 
played a counter strain. This practice was followed by com- 
posers after Weber, and Wagner takes the laurels in dividing 
the strings, in the second act of "Tristan and Isolde/' where 
he has written fifteen separate parts for the violins. To 
Philidor goes the distinction of first writing harmonics for 
the violin when he included this effect in his opera "Tom 
Jones," written in 1765. Berlioz was the first composer to 
attempt writing a full harmony in harmonics, and the effect 
is a heavenly revelation. Paganini, a contemporary of Ber- 
lioz, entered into the competition and reached the ultimate 
in something or other by producing harmonics as high as 
the twelfth series, by using fine, light strings. Although 



not much of a contribution musically, this is quite a feat, 
as can be realized when we consider that Koenig, famous 
acoustician of Paris, was able to reach only six series above 
this, by the use of fine-spun wire. 

Without the violin family we could not have the symphony 
orchestra. This is not to say we could not have a great musical 
organization without the strings, for that is exactly what we 
have in the modern wind band. The wind band, however, is 
distinctly different from the symphony orchestra and always 
will be. It is not far from the truth to say that the strings 
are the symphony orchestra, so important has been their 
part in its glorious development and so characteristic of its 
performance and color have they become. 

Besides sheer agility, the violin has a wealth of technique 
and effects which have made it the valuable instrument it is. 
In addition to tremolo and pizzicato, violinists learned to 
mute the violin by putting a small clip on the bridge, and 
music was marked "con sordino." Novel and useful effects 
were developed by bowing the strings with the wood part 
of the bow instead of with the hair, and music was marked 
"col legno." Bright, crisp effects were discovered by playing 
close to the bridge, and music was marked "sul ponticello." 
Two notes can be played together, and broken chords of 
four notes can be sounded, besides characteristic passages in 
which the notes alternate between the adjacent strings. 
Some of the most beautiful effects in music and colorings 
which cannot be duplicated on any other instrument are the 
delicate harmonics, produced by lightly touching the bowed 
strings at certain spots. Notes can be played "portamento" 
and "giissando" or can be separated by "staccato" or even 
bouncing bow, called "spiccato." Most of these effects are 


possible on the viola, but as the size of the instrument in- 
creases, some of these effects become too difficult or entirely 
impossible on the cello and string bass. 

Since the strings are such an extremely versatile family, 
it is no wonder they have become the great nucleus around 
which the entire orchestra has been built. 



The Flute 


THE FLUTE is the oldest wind instrument known, and its 
development is a story of great inventions, as thrilling 
as the discovery of steam power or radio. The fundamental 
principle of the flute was probably discovered thousands of 
years ago by some Neanderthal or Cro-Magnon man, when 
he noticed that the soft wind blowing across the top of a 
broken reed made a pleasant sound. This was an observation 
requiring intelligence of the highest type in primitive man 
and takes its place alongside of his awareness of numbers 
and of time. 

No doubt centuries passed before our remote ancestors 
rose to an intellectual stature which enabled them to com- 
prehend that low notes came from long reeds and high notes 
from short reeds, and to combine these reeds of varying 
length like the pipes in a pipe organ to make the well-known 
syrinx or pipes of Pan. As far back as we can go in history 
we find the syrinx. The Egyptians played this ancient flute 
long before they wrote history. In their mythology the in- 
vention of the syrinx is credited to Osiris, the great god of 


the underworld. Thus do they prove the existence of the 
syrinx before recorded history. 

Plato was acquainted with the pipes of Pan, and in one 
of his best-known dialogues, "The Republic/' he has Socrates 
engage Glauco, a musician, in conversation about these 
primitive flutes. It is evident Socrates does not think well 
of the pipes, saying they are bad for the morals of the people. 
He amplifies this statement by saying no woman can listen 
to the Lydian pipes and remain virtuous, and he finally con- 
cludes they should be banned from his ideal state. Most 
people who have heard these primitive pipes played seem 
to think Socrates was unduly alarmed. Some people, though, 
conjecture there may have been something to what Socrates 
said, for three hundred years later the notorious Cleopatra 
strode spectacularly across the pages of history and Cleo- 
patra's father was a flute player! This is shown by his name, 
Ptolemy Auletes, which means : Ptolemy, the flute player. 

Hundreds of years after the syrinx was invented another 
great musical discovery was made, quite by accident. All 
primitive peoples are fond of ornament, and a favorite way 
of ornamenting wood articles is by burning designs on them. 
It is only natural that the pipes which made up the syrinx 
should be decorated in this manner. These designs were often 
a series of spots, placed according to the fancy of the primi- 
tive artist. One day, we can imagine, a man sat dreaming in 
the sun and lazily burning spots on his syrinx. Through 
carelessness or through curiosity, he burned the holes com- 
pletely through the walls of the tube. When he tried to blow 
the pipe he found its pitch changed. He also found that by 
covering the holes with his fingers he could still blow the 
original note for which the tube was intended, and that as 



(i) Egyptian pipes of Pan used five thousand years ago. (2) Ancient 
bone flute made from the shinbone of the deer. (3) Chinese cane 
flute. (4) Boxwood flute with four keys, used about 1750. (5) Mod- 
ern Boehm-system flute. (6) Modern Boehm-system piccolo. 


he uncovered each lower hole the pitch ascended step by step. 
Here was a miraculous discovery and one which had far- 
reaching effects on the history of music. 

Although the syrinx survived for centuries, it was finally 
supplanted by this new type of flute. The single-tube flute 
with the holes in its side was more compact, easier to carry, 
easier to make and easier to blow. The notes in the scale 
were accidents, governed at first by designs which seemed 
attractive and later by a spacing of holes which seemed to 
fit the fingers most conveniently. The design on the tubes or 
the convenient placing of the finger holes was of primary 
concern, and what sounds came out of the holes just hap- 
pened. Inspection of many of these primitive flutes shows 
that some of them have holes equidistant apart, while others 
have holes in two groups. Sometimes there are two holes to 
a group, sometimes three or four. Often there is a hole on 
the back side for the thumb, showing definite interest in 
the musical purpose of the flute as distinguished from the 
merely ornamental. J^ 

It is more than a coincidence that there are eight notes in 
our scale and eight fingers on our hands. The scale that could 
be played was composed of those notes coming from the 
eight holes which could be covered conveniently by the 
fingers. If man had been created with only two fingers or 
with eleven, the chances are our scale now would be com- 
posed of a different number of notes. If man's fingers had 
been thicker or longer, the chances are our scale now would 
have different intervals. As it turned out, our ancestors 
became used to those notes which could be sounded by cover- 
ing and uncovering eight holes spaced so the fingers could 
reach them conveniently. 


Some authorities go still further and show the reason why 
there is a semitone between the third and fourth and between 
the seventh and eighth tones in the major diatonic scale. 
This theory has resulted from the inspection of scores of 
primitive flutes. When the two hands are placed along the 
flute, the eight fingers covering the eight holes, the fourth 
finger of the right hand covers the hole from which the lowest 
note is emitted. As each lowest finger is raised, an ascending 
scale is sounded. The intervals are alike except those between 
notes one and two and between notes five and six. It will be 
observed that both of these short intervals lie under the 
third and fourth fingers, and it is believed that these intervals 
were made short in order to accommodate the short fourth 
finger. The first consideration was comfort for the short 
fourth finger, but after this scale was played for thousands of 
years, it became familiar and was judged sweet to the ear. 
In order to prove how at home we have become with the 
conventional scale having the two half-steps, we need only 
try a scale composed entirely of whole steps. These whole- 
step scales sound strange and weird. We expect the half-steps 
to occur in their accustomed places, and when they do not 
appear, we immediately feel that something has gone wrong. 
Not one person in a hundred can sing an octave of whole 
steps without the aid of the piano or some other musical 
instrument. Involuntarily and in spite of ourselves, we drop 
into the established mode of the diatonic scale, with the two 

Of course, in ascending our major diatonic scale, the half- 
steps do not appear between one and two and between five 
and six, but between three and four and between seven and 
eight. Here again we are under the influence of musical cus- 


torn. If we go to the piano and start our scale on "mi" in- 
stead of "do" and play up the scale eight white notes, we 
get the scale which was built into the primitive flutes. Many 
people believe this is the primitive scale from which all others 
in use today were derived. Still other scales were developed 
by starting them on other notes in the primitive scale. The 
Greeks recognized seven of these scales and called them 
Lydian, Phrygian, Dorian, etc., after the peoples who were 
supposed to have originated them. In this classification, the 
Dorian scale is the one which was built into the primitive 
flutes. We have adopted the Phrygian, which happens to 
begin on "do," and we have become so familiar with the 
scale beginning with "do" that what was probably our first 
scale sounds odd to us, as do all other scales beginning on the 
other notes in the scale. 

^The flute with holes in its side is found among all primitive 
peoples in all parts of the globe. Next to the drum, the flute 
is the most common musical instrument of which we know. 
It is made of every conceivable material capable of forming 
a tube through which holes can be pierced. It is made of 
cane, wood, pottery, stone, metal, leather, bones. The Suri- 
nam Indians of Guiana, South America, have the gruesome 
custom of making flutes from the shinbones of their slain 
enemies. The shinbone, or tibia, of animals has always been 
a favorite material for making flutes. In fact, the Latin 
name for "flute" is "tibia." 

Many strange uses are made of the flute by various peoples. 
The ancients used to use the flute to encourage and inspire 
their armies, and the fife, together with the drum, is still 
used for this purpose. The Carib Indian of South America 
has the quaint custom of warning his household of his 


approach home. This is not only a beautiful idea but it has, 
as usual in such matters, a very definite utilitarian value. 
When the man approaches home in the dusk or dark, playing 
a familiar tune, he is spared being mistaken for a hostile 
warrior and is welcomed into the domicile. The North 
American Indian used the flute as a trysting instrument and 
for calling to his sweetheart. The African Kaffir uses the 
flute for calling his cattle. 

Ancient peoples were fond of the flute. It played a promi- 
nent part in the lives of the Egyptians, Hebrews and Greeks. 
The latter developed flute playing to a fine art, as they did 
other arts. Great flute-playing contests were held as a part 
of the athletic games, and there was keen rivalry for the 
flute-playing honors as well as for the athletic. In Europe 
flute playing became popular in the eleventh century, and 
it increased in popularity until, in the sixteenth century, 
it became the vogue even for kings to devote themselves to 
the art. When the notorious Henry VIII died in 1547 he left 
a collection of flutes consisting of seventy-six end-blown 
flutes, seventy-two side-blown flutes and six fifes. 

The fourteenth century marked the parting of the ways 
of flutes blown in two different manners. Some flutes were 
blown across the end and were called flageolets, beak flutes 
and recorders, or English flutes. Other flutes were blown 
from the side and were called transverse, or German, flutes. 
After the fourteenth century, the end-blown flute gave way 
in Europe to the side-blown flute, and today this is the type 
flute with which we are familiar. The only survivor of the 
end-blown flute is the flageolet, or tin whistle, which school- 
boys buy for a dime. In England, however, the end-blown 
flute found greater favor than on the Continent and is often 



called the English flute for this reason. Samuel Pepys in his 
diary has made this flute immortal by his repeated references 
to it and by his praise of its musical qualities. After 1700, 
though, even in England it had to step aside in favor of the 
superior transverse, or German, flute. Bach and Handel 
wrote for both types of flutes, but by the time of Haydn the 
flageolet was no longer found in the orchestra. 

In spite of Pepys' praise of the English flute, or recorder, 
it had serious defects which eventually spelled its doom. 
In principle, the recorder was constructed like the common 
tin whistle or the spring variety of whistle which small boys 
make by slipping the bark from a small sapling when the 
sap begins to run. The mouthpiece constructed on this 
principle is fixed and cannot be controlled as can the reed 
on a clarinet, the lips of the trumpet player, or the air 
stream employed in blowing the transverse flute. For this 
reason, the only method of obtaining the higher notes on 
the instrument was by blowing with more force. When this 
was done, of course the loudness increased, too. It was there- 
fore impossible to play the upper notes softly or the lower 
notes loudly. If the lower notes were played loudly the 
instrument would "overblow" and the higher notes would 
sound. Conversely, if the upper notes were played softly, 
they would fade out and the lower notes only would sound. 
Since it was necessary to play the complete range of the 
instrument both loudly and softly, as the music required, 
the recorder had serious limitations. As musical appreciation 
increased, the defects became more and more evident, until 
the recorder is now only an interesting antique or a plaything 
for children. 

When Striggio and Corteccia made their daring experiment 

in 1565 with their crude orchestra to be used as accompani- 
ment for light plays and embryonic opera, they naturally 
called upon the popular flute to take part. Along with a few 
string instruments and several brass instruments, six flutes 
and flageolets of various sizes were used by these two Italian 
pioneers in orchestra. The flute also took part in another 
venture in orchestra building, when, sixteen years later, 
Balthasarini gave his first performance in France of his 
"Ballet Comique de la Reine." Three flutes were scored for 
in 1600 by Jacopo Peri in his odd assortment of instruments 
used to supply chord accompaniment to singing and dramatic 
reciting in his famous "Euridice" opera. Also in the same 
year Emilio del Cavaliere showed his appreciation of the 
flute by giving it a place as the only wind instrument among 
his otherwise string orchestra, which played in his oratorio 
"La Rappresentazione delP Anima e Corpo," the first orato- 
rio ever produced. His "orchestra" was a sort of ensemble 
consisting of a harpsichord, a double-guitar, a viola da gamba, 
a violin which doubled the soprano voice, and two flutes. 

The "Founder of the Orchestra," Claudio Monteverde, 
seemed to have a poor opinion of the flute, for in his "Orfeo" 
opera orchestra the first orchestra really worthy of the 
name he used only one flute. It seems likely that even this 
one flute was not a full-grown flute, for it is designated as 
an "octave flute," and it is likely that the instrument is 
similar to our piccolo. Later Bach wrote for the piccolo, 
and Handel used it in the aria of "Acis and Galatea" for 
piccolo obbligato. 

When the flute began to take its place in the first orches- 
tras it was by no means the great instrument it is today. 
It was a plain wood tube with finger holes only. The wood 



was usually yellow boxwood, sometimes cocuswood or grena- 
dilla. The better flute was ornamented with rings of ivory 
at the joints and the edge of the bell. It had no keys, and its 
musical possibilities were accordingly limited. Considerable 
improvement had been made in the tone quality and the 
intonation over the first crude flutes fashioned by our 
primitive ancestors, but it was substantially the same in- 
strument invented quite by accident back in the Stone Age. 
With the awakening in music which was taking place in the 
sixteenth and seventeenth centuries came an interest in im- 
proving the musical resources of the instruments; and about 
1677 came the third great invention in the story of the flute 
the invention of a key! This invention played a prominent 
part in the creation of the modern orchestrar^p^ * 

Nobody knows who invented the first key for the flute, 
but he undoubtedly had an ingenious and resourceful mind. 
At some indefinite time keys seem to have appeared for the 
first time on bagpipes, but this fact hardly minimizes the 
achievement of the man who first used the key on the flute. 
Before this great invention the conventional flute had six 
holes from which could be coaxed a more or less satisfactory 
diatonic scale. The lowest note was D above Middle C, 
played with all holes covered, the note coming from the 
end of the tube. As each lowest hole was uncovered, the 
notes E, F, G, A, B and C were sounded. The second octave 
could be played in a similar manner by overblowing. To 
obtain the half-steps needed for the chromatic scale, two 
practices were resorted to, neither of which had much to 
recommend it. If a hole were only partially covered by the 
finger so it leaked air, the resulting note would be about a 
half-step lower than the note which would be sounded when 

the hole was completely covered. Another practice was to 
leave an open hole between covered holes. For instance, if 
the three top holes were covered, G would sound; and if the 
four top holes were covered, F would sound. But if the three 
top holes were covered, the fourth left open, and the fifth 
covered, something resembling F# would sound. This prac- 
tice was known as fork fingering. 


o o o 

o o 

o o I 


Illustrating how fork fingering was used to play the semitones 
on flutes and other woodwinds before keys were invented. When 
the top three holes were covered, G was sounded. When the top 
four holes were covered, F, a whole tone below, was sounded. But 
when the top three holes were covered, the fourth hole left open 
and the fifth covered, F$, a semitone below G, was sounded. 

By these two practices it was possible to obtain the half- 
steps necessary to play the chromatic scale, but these notes 
were weak, "fuzzy," of poor quality and decidedly capricious. 
Flute players of the day prided themselves, though, on the 
perfections of the flute. They really thought it was nearly 
perfect. This is difficult for us to believe, just as it is always 
difficult to believe those things which are strange and un- 
familiar. After a musical standard has been accepted, it is 
difficult to understand how any other could be tolerated, 



just as the old Phrygian scale seems to us the only sensible 
scale, even though the Dorian at one time was accepted as 
perfect by primitive musicians. 

It was a flute of this kind which called out the famous 
remark by Cherubini. An orchestra conductor who had only 
one flute sighed for more and remarked to Cherubini, "What 
is worse than one flute in an orchestra!" The great composer 
quickly remarked, "Two flutes." No two were alike, and 
the disharmony of two flutes must have justified Cherubim's 
caustic remark. 

Not only did the flute players before the invention of the 
key accept this crippled and decrepit scale, but they made 
a great virtue of its very faults. Much depended upon the 
player. Almost any note might come out of any hole. The 
player must favor the note one way or the other to make it 
fit into any scale. This very uncertainty and lack of depend- 
ability of the notes focused the flute player's attention on 
playing the notes sharp or flat to make them in tune. The 
result was that they began to talk of enharmonic differences! 
In a time when it was difficult to play a simple diatonic scale 
in tune and when it was next to impossible to play a chro- 
matic scale, these flute players became obsessed with the idea 
that C# should be sounded sharper than Db, that Bb had 
fewer vibrations than Aft! 

When flute players had such odd ideas about the perfec- 
tions of the flute, it is a wonder that the first key was ac- 
cepted at all Apparently, though, the D# was more difficult 
to obtain than other semitones, and the D# key was an 
improvement which was readily grasped. There had been 
only one good way to play D#, and that was by partially 
covering the E hole. According to William Kincaid, first 


flutist of the Philadelphia Orchestra, it was also possible to 
play D# on the old flute by fingering Etj and stopping the 
end of the flute against the knee, but this produced a D# of 
poor quality. It was impossible to play it by fork fingering, 
for D was played with all holes covered, and the tone came 
out the end of the tube. For fork fingering it would have 
been necessary to close all holes above E, leave E open, 
and close the D hole immediately below it. This was im- 
possible, since there was no D hole. The only way left was 
to compromise the E hole. This had to be done with the 
third and weak finger of the right hand, and since this finger 
is the least agile on the hand, the playing of D# always 
presented a problem. 

With a real flash of genius, some unheralded inventor 
conceived the idea that if he bored a hole between the E 
hole and the end of the tube, he could obtain a semitone 
between E and D. After he had bored the hole he found 
difficulty in covering it. The fourth finger of the right hand 
was short and did a poor job of covering and uncovering it. 
Finally the inventor adopted the key idea. With a hinge as 
a fulcrum, a pad for covering the hole at one end and a 
spatule at the other, within easy reach of the little finger, 
he could open and close the hole with greater ease. A spring 
was placed under the spatule end of the key which raised 
this end in the air and depressed the other end, thereby 
keeping the hole covered. When the player wished to open 
the hole to sound D# he covered the E hole and all other 
holes above and pushed down on the spatule; when he wished 
to close the hole to obtain Dt], he released pressure on the 

As is always true in instances of this kind, we who are 



familiar with an invention or discovery cannot comprehend 
the originality of mind which first conceived it. The key on 
the flute seems to all of us most obvious and nothing to get 
excited about; but when it is realized that thousands of 
musicians for thousands of years did not hit upon this idea, 
the greatness of the achievement begins to dawn upon us. 
For the same reason it is difficult for us to comprehend why 
the Ancients, who invented mathematics, geometry and 
astronomy, and brought to a high stage of perfection the 
arts of poetry, painting and sculpturing, never discovered 
the art of sounding a string instrument by bowing. The 
strange fact remains that they did not. 

In all our thinking, whether about music or anything else, 
we are hedged about by the great, black unknown; and 
usually when original minds pierce the unknown, it is be- 
cause of some new urge which focuses men's attention on 
certain sectors. 

For a hundred years before the first flute key was invented, 
there had existed in Italy particularly, but also in France 
and Germany, a great and new interest in music. Many 
experiments had been conducted with opera and oratorio 
and with instrumental music to go with them. The idea of 
making music a part of drama not an accessory, but an 
integral part was a new idea. Men were intrigued with the 
experiments. The effects found favor with the people. Atten- 
tion was focused in this direction. Whereas for centuries no 
great pressure had been brought to bear upon musicians and 
the makers of musical instruments for something new and 
novel of this kind, now there began a great and intensive 
movement which has finally grown into what we know as 
the classical music of the masters. Bach and Handel were 


the first great fruits of this movement in composers. The 
superb violins of the Italian master craftsmen were the first 
great fruits of this movement in musical instruments. The 
invention of the first flute key by an unknown genius was 
another important fruition of this great musical urge. 

Strange as it may seem, it required of the fertile mind of 
man a half-century more to invent the second key. The 
supposed perfection of the flute of the time kept anyone from 
thinking there could be any further improvements. Now that 
the player could make D# it was assumed that the millen- 
nium of the flute had arrived. 

The greatest flute virtuoso of the day was Johann Joachim 
Quantz, who is famous as the flute instructor of Frederick 
the Great of Prussia and later as court musician during 
Frederick's reign. Imbued with the musical thought that 
enharmonic differences should be recognized, Quantz finally 
came to believe that there was a slight shortcoming in the 
flute which he had been taught to believe was perfection 
itself. Year after year this conviction grew upon him, until 
he finally decided something should be done about it. He 
conceived the idea of a second key, a key which would correct 
the final error in this almost perfect flute. 

And what was this error or shortcoming, and how did 
Quantz overcome it? It is a little difficult to take Quantz 
and his invention seriously, but it seemed a very serious 
matter to Quantz. Accustomed to playing sharps and flats 
with due recognition of their enharmonic differences, Quantz 
could not accept for Eb the note which came out when the 
D$ key was used. This is a classical example of straining at a 
gnat and swallowing a camel That Quantz would pick out 
this difference between Eb and D# as a serious fault the 



only surviving fault on a flute which was so obviously out 
of tune throughout, is an absurdity which is hard for us to 
understand* Quantz could not rest until he had corrected 
this fault, so in 1726 he invented another key, for playing 
Eb. This was placed alongside the D# key, with the hole 
made slightly larger and located a slight distance below that 
of the D# key! 

Few players appreciated this invention; it was not used 
outside of Germany, and but very little within Germany. 
Quantz never seemed to suspect the real reason, but thought 
it presented too great difficulties for the player. In one of his 
instructions for the flute he goes to considerable trouble to 
show that this key was really not so difficult to play as it 
might seem! 

If Quantz's inventive genius went unrewarded, his ability 
as a player was widely acknowledged. It was Quantz who 
first convinced Scarlatti, the composer, that the flute was a 
musical instrument of merit. Someone asked Scarlatti to 
listen to a flute. He replied that he detested wind instruments 
because they were never in tune. After some persuasion he 
finally consented, much against his better judgment. The 
player happened to be Quantz, and Scarlatti had to admit 
the instrument could be played with artistry. He thought so 
well of the flute or rather of Quantz, the flute player that 
he afterward wrote two solos for Quantz. 

Two years after Quantz invented the Eb key he became 
instructor to Frederick, heir to the throne of Prussia. 
Frederick was only eight years old when he began to take 
lessons from Quantz, and he was an apt and talented pupil. 
The old Prussian king became alarmed when he saw his son 
becoming an addict to flute playing. He considered it effem- 

inate and told Frederick and Quantz it would have to stop. 
Frederick, though, was such a lover of the flute that he could 
not give it up. He continued his lessons with Quantz on the 
sly, and on several occasions the flute lesson nearly became 
disastrous for both. Once the old king came upon them by 
surprise, but Frederick saved the day by hiding his tutor 
and his flute in a closet by the fireplace, used for storing 
wood. When Frederick finally became king the loyalty and 
help of Quantz were rewarded. He was made royal court 
musician at a liberal salary, and when he became old he was 
pensioned. Frederick became an accomplished player and 
composer on the flute. He played for court audiences and 
often selected difficult concertos which he played with the 
court orchestra. Although a great monarch, he always was 
conscientious in his playing and wished to be judged by his 
playing alone and not because he was king. He was very 
nervous when appearing in these concerts, and he would 
spend the time before he was to play practising difficult parts. 
The only person he would allow to criticize him, however, 
was Quantz, his old and trusted tutor. Once a member of 
the orchestra had the temerity to offer a criticism, and the 
next day he lost his position. 

Although the Eb key did not long survive, Quantz made 
another improvement which we find on flutes today. This is 
the movable cork in the head joint which makes it easier 
to tune the flute. Another inventor or inventors found that 
the small blowhole on eighteenth-century flutes could be 
improved by changing the shape and size. After considerable 
experimenting it was determined that the tone could be 
produced easier and could be controlled better if the hole 
were oval instead of round and also if it were made larger. 



At various times after this the hole was changed, one time 
to square, but the conclusion of these earlier inventors has 
been substantiated by the fact that the finest flutes today 
still have the large oval blowhole. The hole on many modern 
flutes is more rectangular than oval, the shape being rec- 
tangular with the inside corners rounded. 

The bore of the early flute was also changed about this 
time from cylindrical to slightly conical; that is, the flute 
near the blowhole was about an eighth of an inch larger in 
diameter than it was at the open end. The conical bore was 
supposed to improve the intonation and response of certain 
notes, but it also weakened the notes in the lower part of 
the scale. Finger holes were reduced in size, and flutes were 
made in as many as four joints. Credit for the conical bore 
and small finger holes is sometimes given to Jacques Hotte- 
terre, the French instrument maker, but it is more likely that 
Johann Christoph Denner, celebrated instrument maker of 
Nuremberg, was responsible. 

Forgetting the ludicrous Eb key, the next keys added to 
the flute were G#, Bb and Ft]. These three keys were found 
on flutes by the middle of the century, and Qj was added 
about 1780. Some flute players who had considered the 
flute as perfect without keys began to say that the adding 
of many keys spoiled the tone and intonation. Among these, 
of course, was Quantz. Wendling, the successor of Quantz, 
also talked against keys, being particularly hostile to the 
low C# and Ct). So strong was the influence of these men that 
some of these keys were abandoned temporarily. Pietro 
Grassi Florio, first flute of the Italian Opera Orchestra, 
thought well of these latter keys and revived them. He valued 
them so much he hung a little curtain over the foot joint 


to conceal the keys so other flute players would not discover 
how he had increased the resources of his instrument. 

Quantz, the first great flute player, was born in 1697. 
One hundred years later was born a musical and mechanical 
genius who was destined to transform the flute into one of 
the most nearly perfect of all wind instruments Theobald 
Boehm. When Boehm was a boy he began his study of the 
flute, using a four-key model patterned after one made by 
Karl August Gresner, famous instrument maker of Dresden. 
It had six holes and four closed-type keys for playing D#, 
G#, Bb and Fb|. Before Boehm was twenty-one he had 
mastered the eight-key flute and had become flutist in the 
Royal Bavarian Court Orchestra. Later he distinguished 
himself as soloist in Paris and London, During all this time 
he was intensely interested in the construction of flutes 
and did a lot of experimenting with tuning and the key 

Manufacturers seemed to have decided that the way to 
improve the flute was merely to add keys. Finally a model 
was produced with seventeen holes, eleven |ceys and four 
special levers. But even the eight-key flute was difficult 
enough to make work, and this complicated mechanism 
must have been impossible. The keys were crudely made, 
and many were hinged on a wire run through a hump of 
wood on the body of the instrument and slotted through 
the center for the key. Being unable to get flutes made 
according to his wishes, Boehm finally set up his own factory 
in 1828 and began building flutes as he felt they should be 
built. Boehm was by trade a goldsmith, and his knowledge 
of this craft came in handy for him. He developed a key 
mechanism hinged on small axles pivoted between metal 



posts screwed into the wood. This was a much more positive 
mechanism and gave the smoothest, fastest action known 
at that time. 

Boehm also did a lot of work on the intonation, response 
and tone quality of the flute. The notes in the scale were 
dreadfully out of tune, and only a very skillful musician 
could play a scale with good intonation. Many notes were of 
poor quality, being dead, "fuzzy," colorless. Some notes 
were difficult to play, and only a powerful man with a well- 
trained embouchure (shaping of lips, tongue, etc., for play- 
ing) could bring them out. Other notes boomed out with 
little effort. Running up the scale of the flute of those days 
was just like running up a flight of stairs whose steps are 
all of assorted sizes. First would be a low step, then a high 
one; next a wide step, then a narrow one. To make any speed 
up a flight of steps such as this, one would have to be an 
acrobat. It required much of the dexterity of an acrobat to 
play a flute in Boehm's day. This problem occupied Boehm 
during every minute he could spare from his concert work, 
but he did not accomplish a great deal. 

Then, during one of his trips to London, Boehm chanced 
to hear the great English flutist Nicholson. This experience 
changed Boehm's entire views about the flute and started 
him on the line of work which completely revolutionized this 
instrument. Boehm was especially amazed at the big and 
powerful tone Nicholson was able to produce. When he 
examined Nicholson's flute, he found the holes were very 
large, some of them too large for Boehm's fingers to cover. 
The irregular spacing and size of the holes offended Boehm's 
innate feeling for symmetry and orderliness. He immediately 
decided that he could get nowhere trying to patch up the old 


eight-key flute. Such a flute had too many serious faults. 
He saw he would have to discard entirely the flute as con- 
structed then and start from scratch. The new flute must 
have holes all the same size, and they must be placed scien- 
tifically along the tube according to the exact divisions of 
the scale. 

With this objective in mind he set to work. He first made 
a flute without any holes in it other than the blowhole. 
Then he carefully cut off the end little by little until he had 
cut it to a length which gave him the first note in the scale. 
Measuring very accurately the distance from the embouchure 
(blowhole) to the end of the tube, he determined where the 
hole should be for this note. Having found this measurement, 
he incorporated it in a second flute. After hours and hours 
of such tedious work he finally produced a flute with fourteen 
holes, all of the same size and spaced along the tube for 
accurate intonation. Here was the most nearly accurately 
built chromatic scale that had ever been constructed for the 
flute. So far, so good. 

But Boehm immediately faced a problem that would have 
crushed a less bold or less resourceful inventor. He sat down 
and counted his fingers. He knew without counting that he 
had only ten. In actual playing only nine fingers were avail- 
able, for the thumb of the right hand had to support the 
flute. The problem boiled down to this stubborn question: 
How can I cover fourteen holes with nine fingers? Nobody 
had ever done it before. It seemed to have no solution. To 
Boehm it must have seemed as if he would need some 
mathematician who could work out a new kind of arithmetic, 
for as long as two plus two makes four there was no way of 
covering fourteen holes with nine fingers. To continue think- 


ing there was a way of cheating at numbers to make this 
possible was likely to drive a man crazy. 

One flute inventor of Boehm's time actually did go crazy 
over this problem. Captain Gordon of the Swiss Guards of 
Charles X was a flute player. Equipped with an ingenious 
mind but with little mechanical skill, he started improve- 
ments on the flute in 1826, Four years later, in the revolution 
of 1830, Charles lost his throne. During one of the battles 
in the streets of Paris, Captain Gordon's men became seized 
with panic and in this disorganized condition were massacred. 
Such a terrible experience is said to have affected his mind. 
Having lost his position in the army, Gordon turned his 
mind to the flute. Such a knotty problem was not a mental 
task suitable to a man whose mind had been so upset. After 
many discouraging attempts to develop his ideas of a new 
flute, he completely lost his mental faculties and is said to 
have died in an insane asylum. Ever since Gordon's death 
there has raged a controversy over whether Gordon or 
Boehm was the inventor of the modern flute. Some of their 
ideas and developments were similar, and many claimed that 
Boehm used some of the unfortunate Gordon's ideas, but 
nobody has been able to deny the great contribution Boehm 
made to flute building. 

Boehm faced the unanswerable question of how nine 
fingers could cover fourteen holes and finally found the 
answer! His answer was the famous ring key. This key was 
like other keys, with a pad at one end for covering the hole 
and a spatule at the other end which the finger touched, 
thereby operating the key. It was different, though, in that 
the spatule was really a ring. This ring was situated so that 
when the finger pressed it down, the ring encircled a hole; 


and at the same time that the pad covered a hole, the finger 
on the ring key also covered a hole the hole encircled by 
the ring. With this ingenious device one finger could cover 
two holes, and the unanswerable problem in arithmetic had 
been solved. 

B ft 

How Boehm was enabled to open and close fourteen holes with 
nine fingers, by the use of the ring key. Ring key A and pad key B 
are mounted on the same axle and are sprung open. When the hole 
A is closed with the finger, the ring is also pushed down. As the 
axle turns, the pad key B also closes the hole B. In this way two 
or more holes can be closed with one finger. 

Since it was not always desirable, when closing the lower 
hole, to close the hole encircled by the ring, Boehm made the 
mechanism so the lower hole could be closed without closing 
the one above. This was accomplished by attaching an arm 
to the ring key which articulated with the key below. When 
the ring key was pressed down, the arm swung down onto 
the key below and closed it. The lower key, however, could 
be closed without disturbing the ring key above. This ring- 
key principle revolutionized the entire family of woodwind 
instruments and is at the basis of practically all the key 
mechanism of our modern instruments. Boehm did not invent 
the ring key and never claimed to be the inventor, but he 
unquestionably established its worth by incorporating it 
with other useful ideas on his flute. Reverend Frederick 



Nolan of Ireland invented and patented the ring key in 1808, 
but it lay buried for the most part in the patent office until 
Boehm put it to this ingenious use on his new flute. 

To an acoustically correct chromatic scale and ingenious 
ring keys, Boehm contributed still another feature open 
keys. The first D$ key had been a closed key; that is, it was 
sprung closed with a spring and was opened by pressing 
down on the spatule end of the key. All the other keys which 
were added after the D# key were closed. It is true that in 
1800 Johann Tromlitz, a flutist of Leipzig, had made and 
invented an open key, but it was Boehm who first developed 
the idea successfully. The open key works in a manner 
exactly opposite from that of the closed key. It is mounted 
so it is poised above the hole and the pad covers the hole 
only when the spatule is depressed by the finger. Boehm 
chose open keys because he saw they could be worked faster 
and with greater ease than closed keys, for in order to spring 
a closed key shut tight enough so the hole could not leak, a 
stiff spring must be used. To open the hole the finger must 
overcome the action of the stiff spring. The open key, on the 
other hand, requires a spring stiff enough only to keep it 
from dropping over the hole, and it takes little strength of 
the finger to overcome such a light spring and depress the 
key. Light action meant speed and ease of operation, and 
that was what Boehm was striving for. 

When Boehm had completed his new flute in 1832 he 
played it in concert and amazed his hearers with his perform- 
ance. Boehm was not a stunt player like Nicholson but ex- 
celled rather in musicianship, interpretation, phrasing. But 
his new flute gave him greater technical display as well as 
more finished performance. It was more nearly accurate in 


intonation and had a positive, light and fast mechanism. 
Best of all, it was playable in any signature. Boehm could 
modulate all over the scale with amazing ease, and his hearers 
could hardly believe their ears. 

He took his flute to Paris and to London. During one visit 
to Paris he took his flute to show it to the great composer 
and conductor Rossini. When he arrived at Rossini's quar- 
ters, Rossini was shaving, and Boehm was invited to wait 
in an adjoining room. While Boehm was waiting, he took 
out his flute and began practising scales, arpeggios, skips 
and trills in various keys. Finally he came to the key of Db, 
in which key it was practically impossible to play on the 
old eight-key flute, the flute with which Rossini was familiar 
and whose resources and limitations he understood. Without 
hesitation Boehm launched into difficult arpeggios and trills 
in this key. Soon Rossini came rushing out into the room, 
one side of his face shaved, the other covered with lather, 
and a razor in one hand. He confronted Boehm with the 
incredulous remark, "You can't play a flute in that key." 

Boehm replied quietly, "But I'm playing it." 

"I don't care if you are," Rossini maintained. "It is ab- 
solutely impossible." 

This little anecdote illustrates the general astonishment at 
the new Boehm flute. Although some great flute players of 
the day changed to the new fingering system, many opposed 
it as newfangled. They criticized it on various points, one 
of them being that the lower and higher notes were of poor 
quality. This was a just criticism and one which was not 
remedied until Boehm produced his 1847 flute with a cylin- 
drical bore. But soon after bringing out his first new flute, 
became interested in various other activities, among 


them being the development of a new type of blast furnace 
for smelting iron ore, and during all this time the flute re- 
mained unchanged. 

In 1847 Boehm came out with an even more radical flute. 
He had taken up the study of acoustics and had applied 
some of the laws of sound to the practical building of the 
flute. He established as most efficient that tube which is 
thirty times as long as its diameter. He also found that the 
defective notes in the extreme top and bottom of the scale 
could be improved by making the bore cylindrical rather 
than conical. His new flute was therefore cylindrical, except 
for the head joint, which was shaped somewhat like a parab- 
ola or truncated cone. In the development of this flute 
Boehm made over three hundred experiments, mostly with 
metal tubes. He found that the larger the hole, the more 
nearly the center of the hole approximated the cut-off of a 
plain tube. By making the holes large he was able to build 
a scale even more nearly acoustically perfect. Large holes 
also improved response and gave greater volume of tone. 
He finally finished by making the holes so large they could 
not be covered by the fingers. To overcome this trouble, 
he adopted covered keys, sometimes known as plateau keys. 
Boehm's experiments with the metal tube led him to use 
metal for this flute, the first metal flute ever made. 

Boehm outdid himself on the second new flute. It was 
superior to anything that had ever been built, having fifteen 
holes and twenty-three keys. But it is often overlooked that 
three important inventions of another man greatly helped 
Boehm in making this flute as good as it was. Between the 
time of Boehm's first and second new flutes, August Buffet, 
Jr, instrument maker of Paris, had invented the needle 


spring to replace the conventional flat spring, the clutch to 
replace the articulating arm on Boehm's ring key, and the 
sleeve to permit keys with reverse action being mounted on 
the same shaft. These ingenious inventions speeded up the 
action of the mechanism, made it more compact and efficient, 
and made practical certain other inventions which without 
them would have been impractical. Next to Boehm, Buffet 
probably did more than any other man to make our wood- 
wind instruments what they are today. 

Although our modern flutes are built with the open-key 
system, catalogs on these instruments carry a phrase which 
is more or less confusing. Sometimes flutes may be had with 
"closed G#" or with "open G#," but in America most flutes 
are "closed G#." The introduction of the closed G# key has 
an interesting history. Coche, prominent flute player of 
Paris, objected to Boehm's open G$, because when the little 
finger of the left hand is pressing the key down to close it, 
it is difficult to lift the third finger. In making high E on the 
open system, for instance, the little finger is down, and the 
first and second as well. To play this note it was necessary 
to raise the third finger, a slow and difficult action. Coche 
reasoned that if this key were of the closed type, the little 
finger would not have to be down and the third finger could 
be lifted more easily. In spite of Boehm's opposition to this 
suggestion, the closed Gft has become universal in France and 
America. The only place Boehni's open G$ key is popular is 
in England and Germany. 

In 1851, at the Industrial Exhibition of All Nations, held 
in London, the Boehm flute carried off all the honors. Its 
fame spread all over the world. An American flute made by 
Pfaff of Philadelphia was entered in this exhibit, but it had 


little chance against the product of the master Boehm. 
The Boehm flute was first made in America by Laribee, 
but many were imported from England and France. Two of 
Boehm's own pupils came to America Wehner and Heindl 
and helped popularize the Boehm system here. Among the 
more noted American flutists was the poet Sidney Lanier. 
He prophesied a great future for the flute when he said that 
the twenty first violins of the symphony orchestra would 
soon be counterbalanced by twenty flutes. Today the flute 
duties of the symphony orchestra are handled by one or 
two and sometimes three flutists. 

Few important changes have been made in the flute since 
Boehm finished his 1847 flute; it is substantially the same 
in all respects. We are directly indebted to Boehm for our 
system of fingering, the cylindrical bore, the metal tube, 
and are indirectly indebted to him in numerous other ways 
for his good judgment in selecting and his skill in perfecting 
the suggestions of other inventors. 

*-* Almost all our flutes today are made of metal. At first 
there was much criticism of the metal flute on the score 
that it was inferior in tonal quality to the flute made of 
cocus- or grenadilla wood. Boehm himself seemed a little 
doubtful on this point but finally straddled the issue by say- 
ing much depended on the individual player. When the first 
metal flutes were used in Wagner's Bayreuth orchestra, the 
old master spied them and ordered them out. He said, 
"Those are not flutes they are cannons." It is doubtful if 
Wagner could have known the difference from their tone 
quality alone. Numerous blindfold tests have been made 
with metal and wood flutes, to the complete confusion of 
everybody, showing that few people can tell the difference 

from the sound alone, although it has been proved by delicate 
recording devices that there is some difference. 

Alto and bass flutes have been made but have had no wide 
use. Boehm made himself a bass flute in G, and he declared 
when he was an old man, after playing this flute for many 
years, that it was his favorite instrument. The alto flute 
is used in many modern scores with good effect, especially 
in the works of Stravinsky and Ravel. The little octave flute, 
generally called the piccolo, is really a small edition of the 
Boehm flute, having the same key mechanism and being 
made of metal. Strangely enough, however, the best piccolos 
are conical in bore instead of cylindrical, like the flute. 

Since Cavaliere's oratorios in 1600, the flute has been one 
of the important instruments in orchestra. In the scores of 
this composer, the two flutes were the only wind instruments 
used, all other instruments being strings. The classical com- 
posers since Cavaliere's time limited the number of flutes to 
two, until Haydn called for an extra one. Later, Wagner and 
Verdi made the use of three flutes common, and symphonic 
bands often have as many as five flutes. 
** The flute is called the coloratura soprano of the woodwinds, 
to distinguish it from the other woodwind sopranos: the 
oboe, or lyric soprano, and the clarinet, or dramatic soprano. 
The flute gets this name from its unusual abilities to execute 
highly embellishing and ornamenting passages. 
~ Most of the early composers, and some of the more modern, 
use the flute to double the violins. It is also used to give color 
to the ensemble of woodwinds. Much music has been written 
for the flute, not only that of a harmonic nature, but con- 
certos and other solo parts. The flute is especially brilliant 
as a solo instrument. 


^ In its lower register, the flute is soft and melancholy; in 
the middle register it is peculiarly sweet and clear, the voice 
taking on a liquid quality; while in the upper register it is 
brilliant and hard. Its range of two octaves and part of 
another, and the shades of color in the three registers, make 
it a valuable instrument. 

The piccolo is used for such effects as whistling of the 
wind, to imitate the shrill voice of the fife in military music, 
and for certain satanic portrayals which have won for it the 
title of "imp of the orchestra/' Its voice is generally too high 
and shrill for extensive use, but it is indispensable for certain 

^Besides being one of the sweetest instruments in the band 
or orchestra, the flute is the most agile of the wind instru- 
ments, being surpassed only by the violin in technical capa- 
bilities. Staccato or legato playing, double and triple tonguing, 
trills, arpeggios, runs, skips it seems there is nothing the 
flute cannot do and do with great velocity. This great ability 
is due both to the fact that the acoustics of the flute are 
almost perfectly worked out and also that the key mechanism 
is exceptionally close to the body, extremely light in action 
and lightning fast in response. 



The Double-Reeds 


THERE is A PECULIAR NOTE of sadness and mystery in the 
voices of the oboe, English horn and bassoon which sug- 
gests the far Eastern lands from which they were brought 
nearly a thousand years ago. Although they have been used 
in Europe for eight or ten centuries, they have preserved their 
oriental character and suggest an expatriated people who will 
not forget and who still yearn for the homeland. 

When the Crusaders streamed down through Constanti- 
nople, Venice and Genoa to Palestine and the holy city of 
Jerusalem, they found the double-reed instruments used by 
the Saracens in their worship, in social and military activities 
and in the small details of daily life. These instruments were 
almost as characteristic of the Turks and Arabs as the turban, 
the scimitar and the Moslem prayers. The thin, nasal music 
of these oriental woodwinds hung in the air like the perfume 
in the Sultan's harem, like the call to prayer sounded from 
the towers of the mosques. This music fascinated the rough 
and hearty Crusaders, and they brought the double-reed 



instruments back with them to Europe, along with the silken 
robes, water pipes, jewels and other accessories of the lei- 
surely life of the more cultured Orientals. 

But when the Crusaders discovered the primitive shawms 
with their double-reed mouthpieces, these instruments were 
already thousands of years old. As far back as we can go in 
Egyptian history and even beyond into Egyptian mythology, 
we find the double-reed shawm. Actual specimens of these 
little instruments have been found in the ancient Egyptian 
tombs of the Fourth Dynasty, or about 3700 B.C: They were 
made of a piece of cane with some finger holes in the side and 
a double-reed mouthpiece on the end- Except for the mouth- 
piece, they were much like ancient cane flutes, but the mouth- 
piece was the part which distinguished them from all other 
instruments and which was responsible for their character- 
istic tonal coloring. This mouthpiece was for all the world 
like a soda straw which has become collapsed at one end. By 
forcing a thin stream of air between the two sides of the 
mouthpiece, the player could produce a reedy, nasal sound 
which we have come to associate with the oboe. 

The making of this little mouthpiece was a great discovery 
which has figured in music for over sixty centuries. In all the 
sounds of nature there is nothing quite like it. It has a charac- 
ter all its own. String instruments, brass instruments, per- 
cussion instruments and other woodwind instruments are as 
different from the double-reed instruments as night is differ- 
ent from day. These two thin pieces of reed which are made 
to vibrate together produce a strange mixture of musical 
coloring which has not quite been duplicated in any other 
way. The weird, mysterious, plaintive sound of the shawm 
has come down to us unchanged and unmodified through 


the centuries, and its secret is wrapped up in the tiny mouth- 
piece made of the two thin lips of cane. 

Nobody of course knows just how the double-reed mouth- 
piece was discovered, for its inception is shrouded in the 
unwritten story of the centuries before history, but the soda- 
straw illustration is probably as good and as accurate a way 
to explain its discovery as any. Everyone knows how a soda 
straw will collapse at the end when it becomes soaked with 
water by holding it in the mouth. This is probably just how 
the first shawm mouthpiece was discovered. Some ancient 
Egyptian boy was probably chewing on a piece of cane as he 
tilled the black soil along the Nile or herded his sheep in the 
plains. Gradually the reed became soft at the end and col- 
lapsed. When the boy attempted to open up the tube by 
forcing ^ir through it, the lips of the cane tube were made to 
vibrate and the characteristic reed sound was produced. 
The sound was sweet to the ear, and this little noisemaker 
was used for amusement. Gradually it became more and more 
popular as its nature became better known and as through 
improvement it became easier to blow, tin time it became 
a real musical instrument. 

This is all speculation, but it is a fairly accurate descrip- 
tion of how the shawm mouthpieces were made for centuries. 
A piece of reed was soaked in water until it became soft, after 
which the end was flattened. This end was then scraped and 
made thin, leaving two lips of reed with a fine slit between 
them. This was essentially the double-reed mouthpiece, and 
no doubt at first only one note could be sounded on it. 
Possibly various lengths of cane were used and were left as a 
part of the mouthpiece to serve as a handle by which to hold 
the instrument. At some time or other the rest of the cane 



tube was decorated and ornamented with colored dyes or by 
burning. One eventful day holes were burned clear through 
the sides of the tube, as related in the story of the flute, and a 
full-fledged shawm was created, capable of sounding several 
notes, all of the same oboe character. 

Certain sounds seem to appeal to certain peoples, depend- 
ing upon their national character. The African tribesman 
seems to have a peculiar liking for the deep boom, boom, 
boom of the drum. The Romans seemed to like best the shrill 
martial blast of the trumpet. The Chinese love the tinkling 
sound of cymbals and the plucking of thin wire strings. The 
Turks and Arabs seemed to like best the nasal twang of the 
shawm. They developed a wide variety of cymbals and gongs, 
flutes and trumpets, string and reed instruments, but the 
double-reed shawm was by considerable odds the favorite 
of them all. 

For thousands of years the oriental fakir used the shawm 
to charm snakes in the streets of the cities along the Nile and 
along the Tigris and Euphrates rivers, ancient cradles of 
civilization. In the harems of Baghdad and in the temples 
of Cairo the weird music of the Orient was intoned on the 
double-reeds. Shepherd boys herding their sheep on the hills 
of Palestine whiled away the hours by playing the shepherd's 
pipe with the double-beating reed mouthpiece. As Arabs 
squatted in their white tents and as camel caravans crossed 
the deserts, the thin, nasal whine of the primitive oboe was 
constantly heard. Even as the snake of the fakir became 
captivated by the strains of the shawm and moved from side 
to side in lazy rhythm, so the whole Orient seemed to be held 
in the magic spell of this little musical instrument made from 
a piece of cane. 

The Crusaders who came to Jerusalem to oust the infidel 
and worship in this holy place also fell under the spell of the 
infidel's music and carried the shawms back with them to 
Europe. There may have been double-reed instruments in 
Europe before this, but there is little evidence to support the 
theory. Possibly the Phoenicians, who were supposed to have 
landed on European shores hundreds of years before the time 
of the Crusades, brought shawms with them in their strange 
boats, but there are no records of it. In any event, the use 
of the shawm to any extent in Europe can definitely be 
traced to these adventurers of the Middle Ages who con- 
quered the Saracens, only to be conquered in turn by the 
Saracens' favorite musical instrument. 

It was not long, after the first Crusaders returned to 
Europe, until the shawms spread through France and Ger- 
many. Instead of the dark-skinned, black-eyed shepherd 
boy of the Nile, the shawm was played by the flaxen- 
haired, blue-eyed shepherd boy of Europe. Instead of being 
blown by the snake charmer, watchmen who made the 
rounds of the towns during the reign of Henry III in the 
thirteenth century used shawms called "waeghts" for signal- 
ing. Soon the shawms in Europe graduated from instruments 
of the fields, hills and market place and were adopted in 
musical organizations. In the fourteenth century Edward 
III used three shawms in his royal band. In France in 1588 
shawms and drums took the place of the older fifes and 
drums, and this custom spread soon to Germany and finally 
to England. 

The shawms were made in various sizes, and they were 
known under various names. Sometimes the larger shawms 
were called bombards, and the smaller were called bom- 



bardinos, or little bombards. In Germany the popular name 
was pommers. In Shakespeare's time they were called 
hoboys, or hoeboys; and later they were called hautboys, 
which literally means "high wood," so called because they 
were made of wood and their voices were pitched high. This 
is the name which has evolved into the present English word 
oboe. Besides the soprano hautboys, there were larger 
shawms of various sizes, which eventually developed into 
the English horn, bassoon and double bassoon. Praetorius,who 
wrote a treatise on musical instruments in Europe in 1619, 
describes seven different sizes of shawms in use at that time. 

European soil seemed favorable to the shawms in more 
ways than one. Besides attaining wide popularity all over 
Europe, they multiplied in variety. Instead of trying to bring 
the few sizes of shawms to greater musical efficiency, instru- 
ment makers tried for years to see how many different kinds 
and sizes of shawms they could invent. At various times there 
were as many as a dozen varieties being played in Europe, 
but as the orchestra began to take definite shape, these 
various shawms one by one dropped into oblivion. The 
soprano shawm in C has survived in the present-day oboe. 

Another soprano of the family, called the oboe d'amour, 
pitched in A, a third below, struggled along for several years 
but could not quite make the grade. Bach was a great admirer 
of this instrument and wrote some beautiful music for it, but 
after Bach died it was left without a champion, and it was 
buried with the great master. Several composers since Bach 
have attempted to revive the oboe d'amour, notably Richard 
Strauss in his "Symphonia Domestica" in 1903, but all of 
these attempts have failed. This seems a pity, for the instru- 
ment possessed a sweet, wistful voice. 

Thomas J. Byrne, oboist with the Detroit Symphony, 
regrets the discontinuance of the oboe cPamour and believes 
it should be used in the scores of such composers as Bach and 
Strauss. It is used effectively in the Detroit Symphony, and 
other orchestras using the oboe d'amour are those of Phila- 
delphia, Cleveland, Cincinnati and Chicago. An outstanding 
solo passage for this instrument is found in the original score 
of Ravel's "Bolero," but the music is often arranged for 
English horn where the oboe d'amour is not available. 

There were two altos in the family which fought each other 
for years for a place in the orchestra. These were the oboe da 
caccia and the cor anglais, better known today as the Eng- 
lish horn. Both were built in F and were quite similar in most 
respects. The oboe da caccia was in the field first. Its first 
recognition was obtained through Bach, in his " Passion 
Music," composed in 1723. There seemed to be at this time 
other altos similar to the oboe da caccia, such as the tenoroon 
in F and the bassoon quinte in F, but there is considerable 
controversy about just exactly what they were like. They 
may have been in the contest along with the oboe da caccia, 
but the extent to which they were similar is not at all defi- 
nitely known. Bach's choice of the oboe da caccia over the 
other two would seem to indicate the superiority of the 
former, and for several years this instrument seemed to have 
the alto position in the family fairly well clinched. But in 
1760 an instrument maker in Bergamo named Ferlandis 
invented the English horn, and the oboe da caccia, which had 
easily vanquished the tenoroon and the bassoon quinte, found 
itself faced with a more formidable contender. 

Haydn, Mozart and Beethoven followed the example of 
Bach in the use of the oboe da caccia instead of the English 


horn, and with such champions as these it is a wonder that 
the oboe da caccia did not win over the English horn by an 
easy margin. After Bach died Gluck made the acquaintance 
of the English horn and found a liking for it, as he used it in 
a number of his operas. Apparently, though, there was some 
doubt in Gluck's mind as to whether or not the English horn 
deserved a place in the orchestra, for in his Vienna score of 
"Alceste," published in 1767, he included the English horn, 
but in his Paris score of the same opera, published nine years 
later, the instrument was dropped. It was not a question in 
Gluck's mind about the relative merits of the oboe da caccia 
and the English horn, but rather about the need of either of 
these two alto voices of the double-reeds. Later composers, 
such as Schubert, Schumann and Mendelssohn, wavered 
between the oboe da caccia and the English horn, using some- 
times one and sometimes the other. The question was not 
settled until in the second quarter of the nineteenth century. 
Although the German composers seemed loath to accept the 
English horn over the oboe da caccia, the French and Italian 
composers, such as Meyerbeer, Rossini and Berlioz, quickly 
recognized the superiority of the English horn and admitted 
it to the ranks of the orchestra. The turning point seemed to 
be when Rossini gave the English horn a good part in his 
"William Tell" opera score in 1829, and when Meyerbeer 
called for it in his score of " Robert le Diable" in 1831. Today 
all double-reed alto parts are played by the English horn, 
and the oboe da caccia is known only as an interesting antique. 
It is entirely possible that this battle of a century was as 
much a matter of names as of relative merit, and possibly 
the English horn as we 'know it was actually used when it is 
referred to as the oboe da caccia. Possibly the state of affairs 


between these two alto oboes was similar to that which exists 
today between the two brass sopranos, the cornet and trumpet. 
Trumpet parts in the orchestra are often played by the 
cornet, and cornet parts in the band are often played by the 
trumpet. At least we do know that the English horn has 
been the most unfortunate of all instruments in the matter of 
names. How it ever managed to come down to posterity with 
the name of English horn is still considerable of a mystery. 
As has been pointed out many times, it is neither English in 
origin nor horn in nature. Some authorities say it took its 
name from the fact that it resembled in tone quality an old 
English instrument called the hornpipe. The instrument is 
also known by the name "cor anglais/' which is French for 
"English horn." Some say this name was given to dis- 
tinguish it from the French horn, which was called merely 
"cor" by the French; one instrument was looked upon as a 
horn of English descent, while the other was looked upon as 
a horn of French descent. The point is not justified, for the 
"cor anglais" is not a horn at all. In its early form, the 
English horn was bent on an angle so the player could reach 
the lower keys. This fact has led to another explanation of 
"cor anglais," this explanation arising from the resemblance 
in the French language of the words angle and English; per- 
sons holding to this explanation think of the name as mean- 
ing angle horn and not English horn. But all of this contro- 
versy is really unimportant in the final analysis, for regardless 
of what name it goes by, it will sound just as sad and melan- 

The middle voices of the double-reed family seemed to 
have had the greatest difficulty justifying their existence, 
and they finally failed entirely. There are some early in- 



stances of the use of a baritone oboe in C, an octave below 
the soprano oboe, and Richard Strauss tried over a hundred 
years later to revive this instrument but without success. 
In its revived state it was called a heckelphone, after the 
famous instrument maker Heckel, whom Strauss induced to 
build it. There was also another early baritone sometimes 
called the bassoon quinte in Eb. This instrument seemed to 
be pitched a whole tone below the oboe da caccia. It was 
often classified as a small bassoon rather than a large oboe, 
for it is described sometimes as a bassoon raised in pitch. 
Haydn scored for such an instrument in his "Stabat Mater." 
This instrument managed to survive until the time of Berlioz 
and seems to be the instrument Berlioz had in mind when he 
referred to the bassoon quinte in his Modern Instrumentation 
and Orchestration, published in 1848. He reported at this time 
that it was not a common instrument but was found only 
occasionally. He added, however, that it should be used more 
than it was, because it had more power than the English 
horn in its lower range and would be beneficial in the military 
band. Apparently this bit of advice went unheeded, for little 
is heard of the bassoon quinte after Berlioz. 

The bassoon in C, which plays two octaves below the 
oboe in C, won an accepted place in the orchestra about the 
same time as the oboe. There seemed from the first a need 
for such a bass voice, and there were no other important 
contenders. Although Afranio, a Catholic priest of Ferrara, 
is often said to be the inventor of the bassoon, all he actually 
did was to give the bassoon its modern shape by doubling 
the tube back upon itself, his invention being used first on a 
form of bagpipe. It is this shape, resembling somewhat a 
bundle of sticks or fagots, which gave to the bassoon the 


name "fagott" in German and "fagotto" in Italian. In 
1550, Schnitzer of Nuremberg, a celebrated instrument 
maker of his day, is reported to have made some exception- 
ally fine bassoons, and from this time on down to the present 
the bassoon in C has held the bass position in the double- 
reed family. The double bassoon in C, an octave below the 
bassoon, was created later, and even as early as Bach and 
Handel its place in the orchestra was accepted. 

After this long sifting process of three or four hundred 
years, four double-reeds have been found to be all that are 
needed in the orchestra and band: the soprano oboe in C; 
the alto English horn in F, a fifth below; the bass bassoon in 
C, two octaves below the oboe; and the contrabass bassoon 
in C, an octave below the bassoon. Out of all the shawms 
which were brought by the Crusaders from the Orient across 
the Mediterranean Sea and overland through Constanti- 
nople, and out of all the offshoots from the shawms as im- 
planted in Europe, only these four have found a place in the 
orchestra and band of the present time. It has been a case 
of the survival of the fittest: the useful instruments have won 
for themselves musical immortality, while those less useful 
have dropped into oblivion. 

Confining our attention to these four survivors and for- 
getting the civil war among the double-reeds for a place in 
the orchestra, it is interesting to trace the development of 
the modern double-reed choir from their first entrance into 
the orchestra until the present day. It seems that the oboe 
first attained a place of importance in music in the military 
and open-air bands. Both the English and the French bands 
seem to have adopted them in the fifteenth and sixteenth 
centuries, and although Lully was one of the first composers 



to use the oboe in the orchestra, he used the oboe extensively 
in the military band before he decided he could use it in the 
orchestra. The oboes of the time were not the sweet-toned 
instruments with which we are familiar, and no doubt com- 
posers were more than justified in their doubts about ad- 
mitting them to the orchestra. They have been described as 
coarse and raucous in tone. The reed used was wider and 
thicker than that used today, and the sound produced was 
rough and strident. No more reliable testimony to this fact 
need be cited than that it was an almost universal practice 
to muffle the oboe by stuffing the bell full of wool cotton. 
The oboe was not common in the orchestra until some time 
after the middle of the seventeenth century. For about a 
hundred years following this, however, oboes were very 
popular, being used more than flutes and rivaling the 
trumpets in favor. 

There is some dispute about who was the first composer 
who saw enough musical possibilities in the shawm to admit 
it to the orchestra. The honors seem to go to Balthasarini, 
for he used oboes in his "Ballet Comique de la Reine," per- 
formed in Paris in 1581. Ninety years later Camber t used 
both the oboe and the bassoon in the orchestra for his 
pastoral play "Pomone," and about the same time, or before, 
Lully was using both the oboe and the bassoon freely in his 
opera scores. Besides these two Frenchmen, there were two 
Italians, Scarlatti and Legrenzi, who about the same time 
had found the oboe and bassoon useful in their opera orches- 
tras for giving greater volume and for very simple color 
effects. The former considered the oboe the principal wind 
instrument, and he used the bassoon to strengthen the string 
basses, but it was seldom heard alone. 


( Bach was the first composer to use all three voices of the 
double-reeds the soprano oboe, the alto English horn and 
the bass bassoon. This he did in his "Passion Music," com- 
posed in 1723. It must be admitted, though, that his English 
horn was an early form of double-reed known to Bach as the 
oboe da caccia, the English horn not having been invented 
until 1760, by Ferlandis of Bergamo, as pointed out above. 
Bach wrote some beautiful music for the double-reeds and 
especially for the oboe. In his second "Brandenburg Con- 
certo" the oboe holds a place in the quartet equal to that 
taken by the violin, flute and trumpet. It is amazing, es- 
pecially in the second movement, how much music can be 
obtained from the three weakest voices in the orchestra the 
oboe, flute and violin. 

Handel, Bach's great contemporary, was not content to 
use two oboes and one or two bassoons, as did Bach and as 
was customary; he was particularly fond of oboes and used 
them as freely as he did violins. In fact Handel often marked 
parts in his scores to be played "either by violins or oboes." 
Sometimes he had oboes doubling violins. It is difficult to 
imagine the effect as pleasing, unless the oboes he used were 
very weak. Bassoons were also used in greater numbers than 
they are today. Because Handel seemed to love the oboe and 
bassoon so much, it was entirely fitting that when the hun- 
dredth anniversary of his birth was observed in 1785, the com- 
memoration orchestra used twenty-six oboes, twenty-six 
bassoons and a double-bassoon against forty-eight first 
violins and forty-seven second violins. The use of the double- 
bassoon was appropriate also, for Handel was the first to 
introduce this instrument to the orchestra, in some anthems 
he wrote especially for the coronation of King George II of 



England, in 1727. Handel was not alone in his use of a high 
proportion of double-reeds, for the Electorate Orchestra of 
Dresden in 1750 was comprised of five oboes and five bas- 
soons against six other woodwinds and twenty-five string 
instruments. The extravagant use of the oboe did not last, 
however, and the use of two oboes and two bassoons came 
to be general. 

Joseph Haydn used two oboes in his First Symphony in 
1795, and in his Symphony in D, written in London in the 
same year, he used two oboes and two bassoons, as well as 
a double-bassoon. Two of Mozart's three greatest symphonies 
the Eb Major, the G Minor, and the "Jupiter," all written 
in 1788 use two oboes, and all three use two bassoons. 
Beethoven used substantially the same orchestra as that of 
Mozart. Although later composers made some changes in 
the orchestra, the use of the oboe and bassoon remained 
practically the same. Mozart and Beethoven, for some un- 
known reason, did not use the English horn, and its more 
common use after their time is one important gain made by 
the double-reeds. 

When clarinets were introduced to the orchestra, some 
composers seemed to look upon them as a new kind of wood- 
wind which would take the place of the oboes. When clarinets 
were used, oboes were usually dropped. An example of this 
form of writing is seen in Mozart's early Symphony in Eb. 
Sometimes clarinets were used to reinforce the oboes, but the 
prestige of the long-established oboe is seen in the fact that 
when both forms of woodwinds were played together, the 
oboe always had the more prominent part. After 1800 the 
natures of these two woodwinds were better understood, and 
composers came to take the view held today, that they were 


both needed in the orchestra because of their different natures 
and different tonal colorings. 

Nobody seems to know for sure when the first key was 
added to the oboe, or by whom. After keys were added to the 
flute with such success in the beginning of the eighteenth 
century, makers of the oboe seemed to take their cues and 
added keys to the oboe. The two-keyed oboe was standard 
practically throughout the eighteenth century, although it 
is claimed Hofman of Rastenburg added keys for G$ and 
Bb in 1727, and there are some reports of a low C# key. The 
conventional two-keyed oboe had keys for playing low C 
and D#, while additional chromatic notes could be obtained 
by an ingenious arrangement of double holes. These holes 
were very small and were placed close together. When both 
holes were covered, the diatonic note of the scale would 
sound; but when only one hole was covered, the chromatic 
note of the scale, a half-tone lower, would sound. Before the 
close of the century Delusse of Paris had concerned himself 
with the bore of the oboe and had changed it, improving the 
accuracy of the notes in its scale and the ease of its response. 
In 1800 five-key oboes were not uncommon, and after this 
keys were added rapidly, as in the case of the flute, makers 
seeming to think that if a few keys helped the instrument, 
a lot would make it nearly perfect. Before 1850 some oboes 
had as many as fourteen keys, including the octave key. At 
first the oboe descended to C only, but later the range was 
lowered to Bt| and sometimes to Bb. Berlioz mentions the 
oboe in Bb, in his famous treatise on instrumentation pub- 
lished in 1848, and in Mendelssohn's "Midsummer Night's 
Dream," composed in 1832, the score calls for an oboe which 
descends to Bb. 



Barrett, a well-known oboe player of the day, and Trie- 
bert, oboist and woodwind maker of Paris, collaborated in 
developing the modern oboe mechanism. Although their 
work was done at the same time during which the Boehm 
flute was being developed, the key mechanism Barrett and 
Triebert produced was original and different. Incidentally, 
a genuine Boehm-system oboe was constructed about 1850, 
but it was a complete failure. Some have held that the sys- 
tems are the same because the oboe system uses some Boehm 
ring keys, but the key system created by Barrett and Trie- 
bert is developed along different lines from that of the Boehm 
flute system. Their system is characterized by a great many 
alternate fingerings of various notes in some cases as many 
as three or four which feature makes a number of otherwise 
awkward intervals much easier to play. They also introduced 
the double automatic octave key and in other less important 
points brought the oboe to a high state of development. In 
the latter part of the nineteenth century the wide, heavy 
reed which had always been found on the shawms gave way 
to the thinner, narrower reed, changing the tone from a rather 
coarse quality to a thinner but more pleasing quality. The 
Germans still use a wider reed than the French, but it is 
narrower than the reed found on the old shawms. 

The key system used almost universally today is the Con- 
servatory system, so called because of its connection with the 
Paris Conservatory. Important improvements were made 
to the Barrett and Triebert mechanism by Gillet, who was 
considered the greatest oboist of all times. He was professor 
at the Paris Conservatory for many years, and his great 
ability on the oboe called for changes in the mechanism to 
facilitate his playing. During his regime the oboe was brought 

to a high state of perfection, and few important changes have 
since been made to the oboe he was instrumental in develop- 

As is true of all the woodwinds, the bassoon was used for 
centuries with finger holes only. Some of these holes were 
plugged shut with pegs and were opened or left plugged shut, 
depending upon the key of the music. When makers started 
adding keys to the other woodwind instruments about 1700, 
the bassoon was not excepted. By the middle of the eight- 
eenth century the bassoon commonly had four keys low 
Bb, D, F and Gft. All of these were down on the lower part 
of the instrument and were used for getting some chromatic 
intervals in the bottom of the scale. Before the close of the 
century, four more keys were added, making eight, only one 
of these being on the upper or "wing" joint, while three were 
on the boot joint, three on the butt or long joint, and one 
the octave key was on the bocal or mouthpipe. Karl 
Almenrader, of Cologne, famous player and maker of bas- 
soons during the first half of the nineteenth century, retuned 
various holes, added certain notes in the top of the scale and 
improved the response and quality of tone. Triebert, who 
helped develop the oboe key system, also worked on the 
bassoon, as did Triebert's brother. By the middle of the 
century it was possible to play a chromatic scale from low 
Bb upward three and a half octaves to F. The key mechanism 
consisted of from twelve to sixteen keys, and there were six 
finger holes and two thumb holes besides, not operated by 
keys. Today the modern bassoon has from twenty to twenty- 
two regular keys, besides several miscellaneous special trill 

Afranio of Ferrara gave to the bassoon its bundle-of- 



sticks appearance by doubling it back once upon itself, and 
this shape has survived for four hundred years. The first 
double-bassoons were made like the bassoon, but when 
doubled back on itself only once, the instrument was very 
long and awkward to handle. Therefore, most double- 
bassoons are doubled back twice; that is, instead of there 
being only two sticks in the bundle, as in the bassoon, there 
are four sticks in the bundle. Occasionally a double-bassoon 
of the old type is used in the orchestra, and it is odd to see 
the long bell of the instrument sticking up high above the 
player like a giant bedpost. 

Practically all the important composers since Bach and 
Handel have given a place of importance to the oboe and 
bassoon. They call upon the oboe usually to intone passages 
expressing contentment, and wherever a pastoral or oriental 
theme is introduced it is almost invariably given to the oboe. 
The oboe is adapted to extremely legato playing but can also 
play staccato. Some of the more modern composers have 
called upon the oboe to play rapid arpeggios and octave 
skips, and some gifted oboe players have written for them- 
selves brilliant solos in which the oboe is blithe and gay. The 
majority of composers, however, are content to write for 
the oboe only slow cantabile and sustained-note passages. 

The oboe cannot be played for a long period at one time, 
because such a tiny amount of air can be forced between the 
thin lips of the double-reed that playing the oboe is like 
holding the breath. The composer recognizes this and is 
careful to allow a few bars of rest in the music every so often 
so the oboist can get his breath. Bach sometimes failed to 
observe this restriction and wrote solo passages for the oboe 
which are almost impossible to play. Schumann, who under- 


stood the piano much better than he ever understood the 
instruments of the orchestra, completely forgot the plight 
of the poor oboe player, and in the second of his three ro- 
mances for the oboe he wrote a passage of eighty-four bars 
without a rest! 

Mr Byrne of the Detroit Symphony believes, however, that 
too much emphasis is placed on the breathing difficulties in 
playing the oboe. Modern instruments and training in correct 
breathing have eliminated much of this trouble, so that the 
oboist today can rival the clarinetist and flutist in playing 
extended passages. Oboists take a tip from the oriental snake 
charmer who can play the "Musette," a short double-reed 
instrument of primitive design, for long periods at a time. 
The secret of his ability is that he breathes through his nose 
while playing, and the modern oboist does likewise. 

In the orchestra the oboist often is the highest-paid player 
of them all. One reason for this is that good oboe players are 
scarce, but another reason is that the oboist has an extremely 
hard job. His instrument is one of the most delicate and 
sensitive, both in the key mechanism and in the response of 
the vibrating column of air. The bore of the instrument is 
considerably smaller than a lead pencil up near the mouth- 
piece, and consequently attack and control of the tone are 
difficult. The many alternate fingerings for various notes 
result in a mechanism which is finely and delicately articu- 
lated and likely to go out of adjustment at the slightest 
rough handling. 

Nor do the oboist's troubles end with the mechanism : the 
making of the reed is not only an important part of playing 
the oboe but a task which requires considerable skill It is 
possible to buy finished reeds for the oboe, but a great many 



oboists choose to make their own. They start with the sea- 
soned cane and patiently cut, shape and shave the cane until 
it has just the right thickness and taper. Then they tie it 
to a small piece of metal pipe which fits over the mouth- 
pipe, or bocal, of the oboe. The two pieces of cane must have 
just the thickness and shape required and must be matched 
for resiliency so they will work together properly in pro- 
ducing the capricious tone of the oboe. 

There is a rather widespread superstition that playing the 
oboe will make the performer go crazy. Although there is no 
basis in fact for this belief, it seems a wonder sometimes that 
so many difficulties with which the oboist is faced do not 
sometimes drive him out of his mind. 

The pitch of the oboe is more difficult to change than that 
of the brass instruments, with their tuning slides, or the 
string instruments, whose strings can be tautened or loosened. 
For this reason Handel established the practice of having 
the orchestra tune with the oboe. Since it was so difficult to 
change the pitch of the oboe, it was easier to have all the 
other instruments tune to it. This custom has survived until 
today. If Handel had had clarinets in his orchestra, he might 
have elected them to set the pitch for the other instruments, 
for they are also difficult to put in pitch. 

The oboe is a small instrument, and its tone seems small in 
comparison with that of a trumpet or trombone, but it has 
such a peculiar timbre that two oboes can hold their own 
in a great orchestra. Even in the wind band, two or three 
oboes will stand out definitely above the great power of the 
brass and the other woodwinds. 

In the orchestra one of the oboists usually doubles on the 
English horn. This instrument is in general respects just 

like the oboe except that it is about 50 per cent larger. In the 
band one of the oboists also doubles on the English horn, and 
when the band marches, all the oboists double on drums or 
saxophones or some other instrument. The reason the oboist 
cannot play in the marching band is that the embouchure, 
or control of the reeds by the lips, is so delicate it cannot be 
done while walking. 

The bassoon is even more peculiar than the oboe and 
English horn. The very names which refer to the various 
parts are ludicrous. The mouthpipe is called the bocal, as 
on the oboe, while the first joint is called the wing joint, the 
end opposite from the mouthpipe is called the boot, and the 
next joint is called the butt. The inside of the boot is lined 
with rubber, so the bassoon can be said to wear a rubber 
boot on the inside! To assist the player in holding the 
instrument, it is equipped with a "crutch" for the thumb. 
Several of the holes controlled by the fingers are bored, not 
straight into the body of the instrument, but on a slant. 
This may seem peculiar, but it is done to bring the top of 
the hole nearer the finger. The bottom of the hole determines 
the pitch of the note, and the top of the hole is placed near 
the finger for convenience. Even then, the fingers of smaller 
hands must be spread uncomfortably to reach some of the 

The composers Beethoven and Mendelssohn seemed to 
understand the bassoon best. This instrument was a favorite 
with Beethoven, and he wrote some beautiful passages for 
it. Consciously or unconsciously, though, both of these great 
composers cast it in such roles that ever since it has been 
known as the "clown" of the orchestra. Beethoven in his 
"Pastoral" symphony used it to depict the gyrations of an 



inebriated musician in the village band, and Mendelssohn 
wrote a march for two bassoons which describes the antics 
of two clowns. Other composers have obtained laughable 
and grotesque effects by compelling the large and clumsy 
bassoon to do little dances and capers which would be more 
becoming to the smaller oboe or flute. 

This is not to imply that the bassoon does not have its 
serious and impressive moments. When playing music fitted 
to its nature, it is truly a great and beautiful instrument. 
In its lower register it is solemn and grave and foreboding, 
as in the fourth movement of Berlioz' " Symphonic Fan- 
tastique." In its middle register it is weird and sepulchral in 
character, as in the Meyerbeer's "Robert le Diable," where 
is described the resurrection of the nuns. Its highest notes are 
dramatic, penetrating and not unlike a "human cry of 
agony." Its range of three and a half octaves is unusual for 
a wind instrument, and within this wide range is such a 
variety of colorings that the bassoon has come to be a most 
valuable instrument. 

Some compositions use the bassoon only as a harmony 
instrument, but even in such passages the instrument adds 
touches of color which would be greatly missed. It must have 
been a passage of this kind around which an amusing anec- 
dote about Hans von Billow centers. It is related that the 
famous conductor was rehearsing his orchestra when he noted 
some visitors in the auditorium, much to his annoyance. 
Rather than order them out, he used a little piece of strat- 
egy. He turned to the orchestra and said, "Gentlemen, we'll 
now rehearse the bassoons." He then began solemnly to 
direct thirty-two measures of rests, after which the bassoons 
gave a couple of grunts and lapsed into silence. Von Billow 

[ 107 } 

continued to beat out sixty-four more measures of rests with 
the same precision as if the whole orchestra had been play- 
ing. When he came to the end he looked around but his 
uninvited guests had long ago slunk out! 

Although the bassoon is quite an old instrument, its key 
mechanism is surprisingly undeveloped. Apparently players 
will put up with crude and clumsy keys on the bassoon which 
they will not tolerate on other woodwinds. There can hardly 
be said to be a key system on the bassoon; it is a sort of 
miscellaneous collection of keys of one kind and another. 
The French system for the bassoon is less developed than 
the German, or Heckel, system, and the latter system is 
becoming almost universal in the United States. Berlioz 
dropped a hint about the bassoon which unfortunately has 
not been acted upon. After seeing what the Boehm system 
had done for the flute and clarinet, and to a less degree for 
the oboe, he wrote that, of all instruments, the bassoon 
would benefit most by adopting the Boehm ring-key system. 
Frederic Triebert, brother of the oboe maker, actually con- 
structed some bassoons after the Boehm system, but they 
were not successful. Possibly someday another courageous 
manufacturer will try it again. 

The early oboes, bassoons and English horns were mostly 
made of yellow boxwood and had a few brass keys. As more 
and more keys were added, the key hinges became longer and 
longer. This development spelled the doom of boxwood, for 
while it is hard and close-grained, it has the one weakness of 
warping. When short, single keys were all that were used on 
the body, it did not matter if the body warped a little. But 
when the hinges became long, a slight warping caused the long 
hinges to bind and stop working. Sometimes cocuswood was 


substituted, but grenadilla wood finally became the most uni- 
versally used and is the best material for fine oboes and 
English horns today. Occasionally these instruments are 
made of ebonite, or vulcanized rubber. So far, no one has been 
successful in making an oboe or English horn out of metal. 
German bassoons are usually made of curly maple, stained 
a beautiful, rich red, while French bassoons are usually made 
of rosewood. English bassoons are made of both woods. The 
keys are no longer brass on any of these instruments but 
nickel silver or silver-plated. 

Berlioz laments the fact that the bassoon quinte was not 
more used in the military bands, because he says it had more 
power than the English horn, whose place in the band it 
could have taken to good advantage. This same form of 
reasoning must have prompted Sarrus, a French bandmaster, 
to invent the sarrusophone in 1856. This instrument is made 
of metal but shows its family resemblance to the oboe, Eng- 
lish horn and bassoons by its double-reed mouthpiece. It has 
a shape similar to that of the bassoon, but its fingering 
system is a sort of cross between that of the saxophone and 
the French-system bassoon. An instrument more nearly 
approximating a metal bassoon was made by Stehle of 
Vienna in 1835. This was a contrabassoon made of brass, 
which was later improved by Cerveny of Koniggratz and 
Mabillon of Brussels. Sarrus, however, is credited with this 
family of brass double-reeds because he was first to produce 
the complete family. 

In inventing the sarrusophone, Sarrus evidently hoped to 
combine the characteristic double-reed tonal quality with the 
greater power of the brass, but while he did succeed in pro- 
ducing an instrument with greater power, it is generally 


conceded that he was not able to preserve the beautiful 
quality of tone of the wood double-reeds. Originally there 
were six members of the sarrusophone family: the soprano 
in Bb, the alto in Eb, the tenor in Bb the baritone in Eb, 
the bass in Bb, and the contrabass in Eb, giving a combined 
tonal compass of over five octaves. Although several mem- 
bers of this family are still used in parts of Europe, the only 
one found to any great extent in America is the contrabass 
in Eb. This instrument is used chiefly in the concert band as 
a substitute for the bassoon, especially in open-air concerts, 
and is gradually growing in popularity. 



The Single-Reeds 


UNLIKE THE FLUTE AND OBOE, the clarinet cannot trace 
its ancestry back to the beginning of civilization. 
Apparently the genius of Mesopotamia and Egypt was not 
sufficient to create the ancient chalumeau, father of the 
clarinet, and it remained for the greater genius of the Greeks 
to bring it forth. It is true that the natives of the Nile today 
play the primitive chalumeau, and this type of instrument is 
also found in Palestine and other spots in the East, but these 
instances no doubt show a spreading of Greek culture to 
these spots rather than the existence of the chalumeau in 
ancient times. After the chalumeau appeared in Greece, we 
find it in Persia and India. Chalumeaux also were found in 
the British Isles at an early date among the inhabitants of 
Cornwall, Wales and Ireland, this little instrument being 
called the pibcorn, or pipehorn. 

The chalumeau apparently was never a popular instru- 
ment, and the clarinet into which it developed was one of the 
latecomers into the orchestra. The late discovery of the 
single-reed and its slow development into a great musical 


instrument were due to two main reasons. In the first place, 
the making of the single-reed is rather more complicated than 
the making of the flute blowhole or the double-reed of the 
oboe. The primitive chalumeau mouthpiece is made from a 
short length of cane cut at the septum so one end is closed. 
Back about two inches from the closed end a deep gash is 
cut, and a sliver of cane is partially split away from the body 
of the tube. The septum keeps this sliver from splitting away 
entirely from the body, and at the place where the split stops 
the sliver is thinned. This produces a little tongue of reed 
attached to the cane tube by a piece of thinned and resilient 
cane and constitutes the mouthpiece of the chalumeau. If 
the description of its making seems complicated, the actual 
job of making the mouthpiece is just as complicated as it 
sounds and requires considerable skill. 

This mouthpiece is inserted in one end of a longer tube of 
cane, open at both ends and pierced by finger holes along its 
length. The instrument is played by putting the mouthpiece 
inside the mouth until the lips close over the short length of 
tube beyond the sliver. This swallowing act is one of the 
peculiar parts of playing the chalumeau, noted with particu- 
lar interest by those who see the natives of Eastern countries 
playing these primitive instruments for the first time. The 
mouthpiece is thus enclosed in the oral cavity, which acts 
as a sort of wind box whose only outlet is through the slit in 
the side of the cane tube, out through the open end into the 
body of the instrument. If working just right, the sliver of 
cane vibrates as the air escapes, and produces the chalumeau 

This sliver of cane really opens and shuts as the air escapes, 
and therefore transmits pulsations of air into the body of the 


Instrument. The sliver of cane is sprung away from the body 
when the cut is made, and the thinning process enables the 
tongue of cane to move back and forth more easily. When 
pressure of air is increased in the mouth, the sliver of cane is 
forced down against the body of the tube, shutting off the 
flow of air; but the resilience of the cane overcomes the air 
pressure and causes the sliver to pull away from the body 
and to open up the hole. This fluttering of the reed goes on 
very rapidly and creates the air pulsations which strike the 
ear and are translated by the brain into sound. The pitch of 
the note is governed by the rate of these vibrations, which in 
turn is governed by the size of the sliver of cane, the thickness 
of the thinned part, and the size and length of the tube which 
constitutes the body of the instrument. 

The second reason for the delayed appearance of the 
chalumeau in the orchestra is that it has acoustical principles 
which are different from all other wind instruments. When 
a single-beating reed produces vibrations in a cylindrical 
body, we have what the science of sound calls a "closed" 
pipe. All other instruments, such as the flute, oboe, trumpet, 
etc., are of the "open" pipe variety. Why this is so will be 
explained in the closing chapter of this book. Here it is 
needful to know only enough to understand why the clarinet 
was slow in developing. On all open-pipe instruments it is 
easy to obtain two octaves of diatonic scale. It is done simply 
and in the following manner: all the holes are covered to 
sound the lowest, or fundamental, note, and as each lowest 
hole is uncovered, the diatonic scale is played in ascending 
order, as described more in detail in the story of the flute. 
The scale thus ascends: C, D, E, F, G, A and B all played 
with about the same blowing pressure. When the player is 

ready to play top C, all the holes are again covered, and by 
increasing the blowing pressure and lip pressure the instru- 
ment is "overblown" to the octave C. Then, by uncovering 
each lowest hole and continuing to overblow, the D, E, F, 
etc., of the higher octave can be played. This is easy to do 
and accounts for the wide popularity of open-pipe instru- 

Now the chalumeau cannot be played in this way, because 
it cannot be "overblown" at the octave. No doubt this 
greatly puzzled all primitive musicians and prevented them 
from using the chalumeau as freely as they did the flute and 
shawm. They would start in and uncover each lowest hole in 
turn and play up to B. Here they would try to overblow to 
the octave, but it would not come. The only way for them to 
continue playing on up in the higher octave was to open 
holes nearer the mouthpiece. But there were only eight 
fingers, and that was not enough to carry the scale on up. 
This short and incomplete scale no doubt was in a large 
measure the cause of the disrepute in which the chalumeau 
was held for so many years and accounts for the long delay 
in its acceptance in the orchestra. 

Chalumeaux were found in Europe in the thirteenth, four- 
teenth and fifteenth centuries, but they did not attain any 
popularity until the sixteenth century. Even then they did 
not have the popularity of the flute and shawm. The early 
composers didn't seem to know anything about them, or else 
they were not interested in them as musical instruments. 
There is a great deal of confusion on this question, due to the 
lack of definite terms for various instruments. The terms 
shawm, flute and chalumeau were used carelessly to refer 
indiscriminately to one or all of these different instruments, 


but those who have taken great pains to distinguish among 
these three types of instruments sustain the view that the 
single-reed chalumeau was a neglected instrument. The early 
Italian opera composers did not use the chalumeau in any 
of their compositions, although there were instruments of 
all kinds and descriptions in these loosely thrown together 
ensembles. Monteverde, who composed his famous "Orfeo" 
opera in 1608 and who continued to write until nearly the 
middle of the century, experimenting with all kinds of instru- 
ments and playing them in manners not thought of before, 
never once used the chalumeau. Another adventuresome 
composer was the Frenchman Lully. He tried a number of 
innovations, but he never used the chalumeau. Bach is noted 
for his curiosity about all kinds of instruments, and there are 
few instruments which he did not at one time or other give a 
trial. He even invented new instruments, apparently feeling 
the need of additional voices and colorings. But he never once 
used the chalumeau. 

Handel, Bach's great contemporary, is said to have ex- 
perimented with the chalumeau, although we do not have 
any of his original scores to prove it. Some authorities say 
there is in existence an incomplete overture in which two 
clarinets are used, but this is doubtful. Possibly the impres- 
sion that Handel wrote for the chalumeau was due to the 
presence of clarinet parts in scores of the "Messiah," but 
these were added forty years later by Mozart. Other clarinet 
parts in some of Handel's other compositions were put there 
by Mozart, who could not endure the high clarion parts and 
rewrote them for clarinet. Mendelssohn performed the same 
service for Bach, changing high trumpet parts to clarinet 
parts, as in Bach's Suite No. 3 in D Major. Rameau is also 

said to have used the chalumeau in his pastoral play "Acante 
et Cephise," staged in Paris in 1751, but it is most likely 
that the so-called chalumeau parts were written for the 
clarion and not for the chalumeau, for it has been pointed 
out that not a note is in these parts which cannot be played 
on the clarion of the time. Here, again, it would seem as if 
the parts for chalumeau, as found in the scores, were origi- 
nally clarion parts which were later played by the chalumeau. 
An even less well-established contention is the statement that 
Rameau used the chalumeau in "Zoroastre" in 1749. Neither 
the score as published that year nor the 1756 score contains 
any parts for the chalumeau, and we may conclude that none 
was written. 

To be entirely ignored by such composers as Monteverde, 
Lully, Bach and Handel is almost enough to raise the ques- 
tion as to the very existence of the chalumeau. These men 
were noted for their willingness to try any instrument. No 
instrument, apparently, was too primitive and crude to be 
given a trial. They wrote for all kinds of lutes, organs, 
guitars, viols and other instruments now obsolete. They even 
wrote for the cornetto and the serpent, instruments which 
we know were very badly out of tune and very poor in tone 
quality. But they did not think sufficiently well of the chalu- 
meau to write a note of music for it. This makes the chalu- 
meau the ugly duckling of the wind-instrument family. 

There were at least two men at this time who had faith in 
the chalumeau, and the fact that they actually wrote parts 
for it proves its existence. These two men were the German 
composer Keiser, who called for the chalumeau in his opera 
"Croesus," and the Italian composer Bononcini, who wrote 
parts for the chalumeau in his opera "Turno Aricino." It 



is a coincidence that both of these men wrote their history- 
making parts in 1710. The ugly duckling obtained definite 
recognition, and although for nearly one hundred years it 
was to take a place in the orchestra subordinate to that of 
the flute and oboe, eventually it was to share equal honors 
in the orchestra with these two instruments and to rise above 
them to become the most important woodwind in the concert 

Johann Christoph Denner, instrument maker of Nurem- 
berg, is usually given the credit for the invention of the 
clarinet in 1690. Although his invention consisted of little 
more than an improvement on the old chalumeau, he no less 
deserves the honor for which he is noted. Denner was a maker 
of flutes, oboes and bassoons and was a skillful and resource- 
ful man. He took up the old riddle of the chalumeau and tried 
to find out why it would not "overblow" to the octave as did 
the flute, oboe and bassoon. He figured that if he could find 
out how to extend the compass of the instrument upward he 
could make it a valuable instrument. It had a tone quality 
which was different from that of either the flute or oboe, a 
tone quality which was equally as beautiful in its own way. 
He no doubt felt it was a shame to ignore an instrument with 
such possibilities. 

Denner soon determined that this instrument did not over- 
blow to the octave but to the twelfth; that is, when all the 
finger holes were closed and the lowest note, or the funda- 
mental, was sounded, the next note to be obtained by in- 
creasing the blowing pressure and lip pressure was the twelfth 
note above instead of the eighth. He also discovered that this 
note came a lot easier if he bored a little hole up toward the 
mouthpiece. When this "speaker" hole was opened, the 

twelfth was overblown with greater ease. He no doubt was 
elated with this discovery, for it opened up to him some of 
the secrets of the chalumeau. But he was immediately beset 
with other practical problems. He could see that the over- 
blown twelfth note would give him an increased range in the 
top of the instrument, but he did not see how he could bridge 
the gap in the scale between this top series of notes and the 
top of the then-known scale. This was a sticker. 

After much experimenting he finally built a chalumeau 
which had a scale of one octave from G below the treble clef 
to G in the clef, to which he added notes at the bottom and 
at the top in an ingenious way. This octave G to G was 
played by covering and uncovering eight finger holes, just 
as if the instrument were an open-pipe instrument like the 
flute or oboe. Below the G was the bell note F, played with 
all fingers down. This gave Denner a series of nine notes, from 
F to G, but he still had a gap between his top G and the C 
obtained by overblowing. How could he connect his scale of 
nine notes with the new series of upper notes beginning with 
the overblown C? In the top of the scale above the eight 
finger holes Denner added two keys which were sprung shut 
over two holes drilled opposite each other. Apparently one 
of these keys was operated by the first finger of the left hand 
and the other on the back side by the thumb. When one key 
was opened, A was obtained, and when both were opened 
together, B was sounded. This gave Denner eleven notes, but 
the top key, or B key, gave him something else. This key also 
acted as a "speaker" key, and when it was opened (all other 
* holes being closed), C and a whole series of notes above could 
be played with some of the same fingerings as those used on 
the notes in the lower series. Thus the baffling gap in the 


scale of this closed-pipe instrument was bridged, giving a 
scale from F below the treble clef to C in the treble clef and 
several notes higher, possibly to F in the top line, treble clef, 
or two complete octaves. 


/ ATB 
^ S& 

Drawing showing how Denner solved the problem of the chalu- 
meau scale. Nine notes, from F below the staff to G in the staff, 
were played by eight finger holes and the "bell" note F. Notes 
from C in the staff upward were based on the overblown twelfth, 
or C, the individual notes being played by the finger holes. The 
stroke of genius was in obtaining A and B in the staff. These two 
notes he obtained by two keys covering holes located opposite each 
other. By opening one he obtained A; by opening both he obtained 
B. The B key also served as a "speaker" key to assist in over- 
blowing to the twelfth. 

Later J. Denner, no doubt a son of J. C., lengthened the 
clarinet and added a key at the bottom of the chalumeau for 
obtaining low F, low E coming from the bell with all holes 
closed. This made the first series of notes from E below the 
treble clef to Bb in the staff, with the second series of notes 
beginning on Blq. This was the famous three-keyed chalumeau 
known during the first half of the eighteenth century. The 
next two keys added were those for low F# and G$. Credit 
for these two keys is usually given to Barthold Fritz of Bruns- 
wick, and since it is known he died in 1766, these keys were 
in existence sometime before this date. This chalumeau with 
five keys is no doubt the instrument with which Haydn, 
Mozart and Beethoven were familiar. 

It is not known when the chalumeaux first were called 
clarinets. This has led to considerable confusion. One instance 
concerns the exact time when composers ceased to write for 

the chalumeau and began writing for the clarinet. The sane 
way to look at this question is to allow for some confusion 
in names which naturally would exist during such a transi- 
tion. Some people hold that immediately after Denner made 
his improvements to the chalumeau, this instrument became 
known as the clarinet. When they see clarinet parts in a score 
written after 1690, they immediately conclude that the com- 
poser had in mind the clarinet. The parts written in 1710 by 
Keiser and Bononcini, they say, were for clarinets and not 
for chalumeaux. 

It is more likely that when chalumeaux are indicated in 
various scores, actually Denner's clarinets were used, for no 
doubt the chalumeau as improved by Denner was for some 
time called a chalumeau. 

In 1751 or 1752, Haydn wrote his first Mass, and he called 
for two instruments which we believe to have been chalu- 
meaux, but we are not sure. Men who have made a serious 
study of this point say that the first undoubted reference to 
clarinets in musical scores was in 1762, when Thomas Arne 
scored for clarinets in his "Artaxerxes." A year later Johann 
Christian Bach, son of the great Johann Sebastian, used a 
clarinet in his "Orione," performed in London in 1763. 
Gluck is said to have scored for the chalumeau in 1762, when 
he wrote his Vienna score of "Orfeo." This chalumeau was 
said to have had eight finger holes and two holes for the 
thumbs. When GlucFs Paris score of "Orfeo" was published 
in 1774, however, clarinets are indicated instead of chalu- 
meaux. The whole question of names is not of importance, 
since the clarinet after ail was only an improved chalumeau 
and not a different instrument. 

The name clarinet was given to the chalumeau because in 



tone it resembled somewhat a small clarion; hence clarionette, 
afterward shortened to clarinet, although the English still 
spell it "clarionet/ 5 The small clarion was high and shrill 
of voice, and the upper part of the chalumeau scale was Hke7 
wise high and shrill; this resemblance suggested the word 
clarinet. Mozart must have helped establish the clarinet 
name, because he rewrote clarion parts for the clarinet. 
Handel's music was full of clarion parts, for the clarion was 
at its height during Handel's time, but Mozart could not 
endure the shrill voice of the clarion and adapted many of 
these clarion passages to the clarinet. This direct comparison 
must have helped in saddling the clarinet name upon the 
chalumeau, and probably marked the time when the chalu- 
meau became generally known as the clarinet. The word 
chalumeau, however, has been preserved by naming the 
lowest register of the clarinet the chalumeau register. 

Mozart showed the world how to use the clarinet in the 
orchestra, although Haydn and Gluck and others had been 
using clarinets for some time before Mozart. Gluck was not 
certain of himself in handling this new voice of the orchestra 
and at first considered it as a kind of substitute for the oboe 
and flute, or at least only a supplement or reinforcement for 
them. He hesitated to use the clarinet as a solo instrument 
as he did the flute and oboe, and confined himself to using 
it as a harmony instrument. Haydn had two clarinets in his 
Eisenstadt orchestra from 1776 to 1778, but he did not 
understand their possibilities until some years later, after 
Mozart had demonstrated their great resources. Mozart was 
attracted by the clarinet when only a child, having written 
a symphony for strings, bassoons and clarinets in 1765 while 
visiting in London. This fact raises an interesting and 


plausible conjecture that Mozart's interest in the clarinet 
was inspired by hearing Arne's "Artaxerxes," performed for 
the first time in London in 1762, and J. C. Bach's "Orione," 
performed for the first time in London in 1763. When Mozart 
visited the great orchestra at Mannheim in 1777, he was 
already familiar with the clarinet, and if he did not learn 
anything about its actual use from the Mannheim orchestra, 
he was to demonstrate eleven years later in his Eb symphony, 
composed in 1788, that the clarinet was a musical instrument 
of the greatest promise. This symphony is often referred to 
as the clarinet symphony because of the prominence given 
to the clarinet. In it he even excluded the oboe entirely in 
favor of the clarinet. This was a common practice for some 
years, but after 1800 the proper place of both instruments 
was well determined. Five of Mozart's symphonies contain 
parts for clarinet, as do the majority of Mozart's operas. 
Haydn, who taught Mozart many things but learned from 
Mozart how to write for the clarinet and other woodwinds, 
included the clarinet in several of his later symphonies and 
in many of his operas. 

Beethoven used the same orchestra in his first symphony, 
the C Major, in 1800, and in his second symphony, the D 
Major, in 1803; and this orchestra included two clarinets. 
In the first of these compositions, however, the clarinet was 
used as a harmony instrument or to add to the volume in 
tutti passages, all the important solo passages being given 
to the flute and oboe. In the second composition, Beethoven 
used the clarinet with the bassoon but showed he had not yet 
become acquainted with the resources of the instrument. In 
his famous "Eroica" symphony, written in 1804, the clarinet 
came into its own, being used to state important themes and 



(i) Razhok, primitive chalumeau from Russia. (2) Zooma'rah, 
primitive chalumeau from Egypt. (3) Urgun, primitive Syrian 
chalumeau. (4) Boxwood clarinet with five brass keys, used about 
1800. (5) Modern Boehm-system soprano clarinet. (6) Modern 
Boehm-system alto clarinet in Eb- (7) Modern Boehrn-system 
bass clarinet in Bb- 


to take solo parts, just as the long-established flute and oboe. 
From this time on, Beethoven wrote hardly anything in 
which he did not give the clarinet an important role. He was 
partial to the upper register and seemed to avoid the lowest, 
or chalumeau, register. Mendelssohn, on the other hand, 
seemed to avoid the upper register and was partial to the 
chalumeau register. Good examples of this type of writing 
are seen in the opening notes of "Elijah" and in the introduc- 
tion to his "Scotch" symphony. Schubert used the clarinet 
often and was evidently fond of it. The same was true of 
Schumann. Weber shows an unusually keen knowledge of * 
the resources of the clarinet, in the supernatural music of 
"Der Freischutz" and in several of his concertos with 
orchestra. Critics say nothing finer has ever been written for 
clarinet than these early concertos, some of them composed 
as early as 1811. After the first quarter of the nineteenth 
century the place of the clarinet was clearly understood and 
firmly established. 

Although the great German composers wrote mostly for 
the three-key and five-key clarinets, there were developments 
going on in France which were destined to give the clarinet 
chromatic resources of which these great old masters scarcely 
dreamed. Some of these inventions happened during the 
lifetime of Haydn and Beethoven, but news did not travel 
fast in those days, and inventions were only slowly adopted. 
Keys were added from time to time until in 1810 Ivan 
Muller, great clarinet virtuoso of Paris, brought out his 
celebrated thirteen-key wonder. The technical resources of 
the instrument were increased by the more elaborate mecha- 
nism. Notable among the various keys contributed by Muller 
were the key for covering the F-C hole, and several trill keys. 

The tone holes were also relocated, and the number increased 
to twenty-two. Although this feature improved the accuracy 
in the scale, It harmed the tone quality, for it is a well- 
established fact that the more holes there are in the body of 
the instrument, the poorer the resonance and tone quality. 

Apparently other inventors recognized this principle, for 
after this thirteen-key model of Muller's, other inventors 
came out with improved models having fewer keys. All of 
these so-called improved models used the basic Muller sys- 
tem. Among such clarinets were the Gassner five-key model 
of 1849, the Kastner six-key and nine-key clarinets of 1848, 
and the Von Gontershausen thirteen-key model of 1855. 
This basic system survived for many years and is still found 
on clarinets used in some military bands. It made its last 
stand in the Albert system clarinet but is rapidly being sup- 
planted by a superior system, known as the Boehm system, 
invented in 1843 by Klose, 

Klose had the euphonious name of Hyacinthe Eleonore 
Klose, and he was one of the greatest clarinet soloists of his 
day. Striving always for better performance, he eventually 
saw that his instrument stood in the way of greater achieve- 
ment. The twenty-two-hole clarinet of Muller lacked reso- 
nance, and even the thirteen keys did not give facility of 
playing in certain key signatures. Manufacturers of the day 
did not seem equal to the job of improving the instrument, 
so Klose added to his role of clarinet player that of clarinet 
inventor. He had seen what great improvements the Boehm 
system had made on the flute, and he had an idea that this 
system would improve the clarinet. He no doubt found con- 
siderable opposition to his ideas, because the Boehm system 
was created expressly for the flute, which is an open-pipe 


instrument and overblows to the octave, whereas the clarinet 
is a closed-pipe instrument and overblows to the twelfth. 
Wise men said it couldn't be done. They spoke with great 
decisiveness about the theoretical principles which precluded 
it. To make a chromatic instrument playable in all keys was 
difficult enough on an open-pipe instrument; on a closed- 
pipe instrument it was impossible. There were already 
twenty-two holes in the Muller clarinet; there would have 
to be many more holes in the Boehm-system clarinet. There 
were even some ^objections to the fourteen holes necessary 
for the Boehm-system flute because of the harmful effect 
on tone. To adapt the Boehm system to the clarinet would 
require many more holes and a mechanism that would be so 
complicated nobody would be able to play it. 

Klose still believed in his idea, and although he was not a 
mechanic himself he decided such a clarinet as he had in 
mind should be built. He finally went to August Buffet, Jr, 
famous woodwind maker in Paris. Buffet had already gained 
fame for his inventions on the flute. In 1835 he had invented 
the needle spring. Later he invented the "clutch" for articu- 
lating ring keys together, and the "sleeve" which permitted 
reverse-acting keys to be mounted on the same axle. Buffet 
listened to Klose's ideas and finally built a clarinet. It was 
a great success, much to the surprise of the know-it-alls who 
said it couldn't be done. Instead of having more holes than 
the twenty-two on the Muller thirteen-key clarinet, it had 
only eighteen holes! The ingenuity of construction is even 
more remarkable when it is considered that only eighteen 
holes were required for the twenty-three keys. The clarinet 
was now playable in all signatures. Many trills before im- 
practical or impossible were now playable with ease. Certain 

bad spots on the old clarinet, such as bridging registers, were 
made much easier. Klose traveled over Europe with his new 
invention and astonished everybody with the great technical 
display made possible by the improved mechanism. 

The mouthpiece of the clarinet has a story of development 
all its own. Although it retained the old principle of the 
single-beating reed found on the original chalumeau, it was 
refined and made more efficient. As the chalumeau and 
clarinet entered the orchestra, they were built in different 
keys. The change of keys was made possible by barrel joints 
of different lengths. The clarinet was made in five pieces: 
bell, bottom, top, barrel and mouthpiece. When a player 
bought a clarinet in those days, it came with two or three 
barrel joints which were interchangeable and which put the 
instrument in different keys, but all barrel joints used the 
same mouthpiece. Some of the mouthpieces on these old 
clarinets are very crude. They were whittled from a block 
of wood or were made from two blocks of wood held together 
by two pieces of metal tacked along the side. They had al- 
ready assumed the shape of the modern mouthpiece and 
used a separate reed, which was tied to the mouthpiece with 
string. About the time Muller brought out his thirteen-key 
wonder, it became the custom to hold the reed on the mouth- 
piece with a metal band, or ligature, which could be clamped 
tightly to the mouthpiece with a screw. The reed was held 
on top of the mouthpiece and was controlled by the upper 
lip. In France a new school of players sprang up which advo- 
cated the reed on the bottom so it could be controlled better 
by the lower lip. This style of playing spread to other parts 
of the Continent and to America, but the old school of play- 
ing persisted in Italy and is sometimes called the Italian 



school to distinguish it from the French school. In America 
almost all players use the reed on the bottom. 

Hardly any other instrument has been made in as many 
keys as the clarinet, unless it is the French horn. Along about 
1800 there were in use as many as twenty different keys of 
clarinets. Denner started the tendency by making clarinets 
in both C and high A. Johann Christian Bach called for 
clarinets in Bb and D. Mozart wrote for clarinets in low A, 
Bb, B and C. There were also clarinets in high E and F, 
although they were more important in the band than in the 
orchestra. As if these were not enough, an instrument maker 
called Horn, of Passau in Bavaria, invented the basset horn 
in F, in 1770. Thus was the alto member of the family intro- 
duced. It was improved by Lotzof Pressburgin 1782 and also 
by Ivan Muller in 1812. Mozart was the first to use it, scoring 
for the instrument in his "Magic Flute" and also in his 
"Requiem." Beethoven became convinced of its merit and 
used it with good effect in his "Prometheus" overture. 
Mendelssohn followed by calling for the basset horn in his 
"Scotch" symphony. Soon after the middle of the nineteenth 
century, the old mechanism on the basset horn was sup- 
planted by the Boehm system, the pitch of the instrument 
was changed to Eb, and it became known as the alto clarinet. 

The bass clarinet is generally credited to Heinrich Gresner 
of Dresden. The date of the invention was 1793, but appar- 
ently no composer used the instrument until 1836, when 
Meyerbeer introduced it in his "Huguenots" opera, in 
Act V. Dumas, goldsmith to Napoleon, is said to have made 
a bass clarinet in 1805 and to have presented it to the Paris 
Conservatory. The bass clarinet used by Meyerbeer was an 
improved model made by Antoine (Adolphe) Sax, In 1838 

Sax brought out a still better model with twenty-two keys, 
which is praised by Berlioz in his Modern Instrumentation and 
Orchestration for its "perfect precision of intonation, an 
equalized temperament throughout the chromatic scale, and 
a greater intensity of tone." These clarinets were made in 
Bb-and were straight like the soprano, so that the player 
had to stand up to play them. The bell reached nearly to the 
floor, with the result that the sound was muffled. Sax remedied 
this defect by making a concave reflector which was placed 
directly beneath the bell and which served to deflect the 
tone upward with greater volume. Later this defect was 
remedied by curving the instrument into the shape of a 
letter S. Sax also made double-bass clarinets in Eb and BBb. 
His double-bass clarinet in BBb was a failure, though, and it 
remained for M. Besson to bring out a satisfactory instru- 
ment of this kind, in 1891. Although Meyerbeer was the 
first to use the bass clarinet, Wagner was the first to show 
what really could be done with the instrument, and since 
Wagner it has been found in practically all opera music and 
in most symphonies. 

The clarinet family today consists of the Bb soprano, the 
Eb alto and the Bb bass. In various symphony and opera 
compositions the A soprano clarinet is called for, and, less 
often, the C soprano. In military bands the Eb clarinet is 
used, but only occasionally is it found in the orchestra, its 
voice being very shrill. Berlioz once used it in the finale of 
his "Symphonie Fantastique." While the Eb and the C 
sopranos are shrill, the Bb soprano is brilliant in tone without 
that piercing quality which has caused many to dislike the 
higher-pitched instruments. The A soprano is pitched a half- 
tone lower, and there are many who believe it has a more 


mellow and richer quality for this reason. If the actual facts 
were known, no doubt this so-called superior quality would 
be found in the mind only. When the old masters wrote for 
A clarinet, there may have been greater difference than there 
is now, due not so much to the half-tone difference in pitch 
but to the difference in perfection of the two instruments. 
Conductors who demand the A clarinet just because the 
original scores called for it have been embarrassed many 
times when they could not distinguish the A clarinet and the 
Bb clarinet apart. A classical example is the case of Meyer- 
beer, when he was conducting his own operas at Stuttgart. 
He found the clarinet player using the Bb instrument, 
whereas Meyerbeer had written the part for the A clarinet. 
Being one of those who are sticklers for the A clarinet and 
believing the A clarinet possessed a tone quality which could 
not be produced by the Bb clarinet, Meyerbeer insisted that 
the clarinet player use the A instrument. The clarinet player 
laid his Bb clarinet down, fumbled around a bit, and then 
picked the Bb clarinet up again. He then proceeded to blow 
through the instrument to warm it up a bit so it would be up 
to pitch and then nodded to Meyerbeer that he was ready. 
After the A passage had been played (on the Bb instrument!) 
Meyerbeer beamed at the orchestra and said, "There, gentle- 
men, that is the color I had in mind." 

It is not only the conductors who insist that parts origi- 
nally written for the A clarinet be played on the A clarinet: 
many of the old-time clarinetists believe the same. When 
they buy a Bb clarinet, they buy a pair a "matched " pair 
one in Bb and another in A. Many other fine clarinetists, 
though, use only the Bb clarinet, playing all clarinet parts 
on this instrument by transposing them. Berlioz objected 


strenuously to this practice, and this has led some to take 
the same stand; but many of these persons misinterpret 
Berlioz* objection. He points out that the lowest note on the 
A clarinet is the low E, which sounds C#. The lowest note 
on the Bb clarinet is also low E, but on the Bb clarinet, due to 
its pitch, this note sounds a half-tone higher than C$ or D. 
If the Bb clarinet cannot play this Qf in the lower octave 
but has to play it in the octave above, the music suffers. 
Few will dispute this point, but the resourceful Antoine Sax 
changed all this when, a little before the middle of the nine- 
teenth century, he added to the Bb clarinet a low Eb key. 
With this key the player of the Bb clarinet can transpose 
any A-clarinet parts and play them as the composer wished 
them to sound. The objection of Berlioz is overcome, for the 
Bb clarinet player can now play the low Qf with ease. Berlioz, 
probably, would find no objection to transposing A-clarinet 
parts on the Bb clarinet, since his objection has been met. 

While the bass clarinet is used in both band and orchestra, 
the alto clarinet is more of a band instrument than it is an 
orchestra instrument. This is probably due to the fact that 
in the band the clarinet is the most important instrument. 
It holds a place in the band of the same relative importance 
as does the violin in the orchestra. As the violin family is 
complete in the orchestra, with violins, violas, cellos and 
basses, the clarinet family is complete in the band, with 
soprano clarinets, alto clarinets and bass clarinets. As the 
most important instrument in the orchestra has a complete 
choir, so the most important instrument in the symphonic 
band has a complete choir. 

The clarinet is often called the dramatic soprano of the 
woodwinds, to distinguish it from the other two woodwind 


sopranos: the coloratura flute and the lyric oboe. In its low- 
est, or chalumeau, register, its tone is sonorous and mellow 
but a little hollow and nasal. In its intermediate register its 
tone is somewhat weak but is effective in soft passages. 
These notes from F to Bb in the staff are the famous throat 
notes which no manufacturer has been able to bring quite 
up to the excellence of the rest of the scale. The middle 
register is the best part of the scale. Here the tone is clear, 
sweet and strong, penetrative and expressive. This part of 
the scale is capable of the greatest " crescendo al diminuendo" 
to be found on any woodwind. The upper register is naturally 
brilliant and sometimes is inclined to have a "reedy" quality. 

Few wind instruments have such a wide variety of tonal 
colorings as the clarinet, and none except the flute can sur- 
pass it in agility. Flowing scales and arpeggios, trills and 
tremolos, legato and staccato playing come easy on the 
clarinet. In fact, there is scarcely anything which cannot be 
played on the clarinet, its possibilities being limited by the 
player more than by the technical resources of the instru- 
ment. While this is particularly true of the soprano clarinets, 
it is also true of the alto and bass clarinets in a degree, de- 
pending upon the size of the instrument. The reed and keys 
on the alto and bass are larger and respond less quickly and 
easily, but the key system is the same on all sizes of clarinets, 
and the player of one can pick up any of the others and play 
them with little trouble. 

In the symphony orchestra, two or three soprano clarinets 
are used and one or two bass clarinets. Often an alto clarinet 
and a double-bass clarinet are called for. In the band many 
more than this are used. In a band of seventy-two pieces, 
from 25 to 40 per cent of the instruments will be clarinets. 

There usually will be twenty-four Bb sopranos, one or two 
Eb sopranos, from two to four altos and a pair of bass clari- 
nets, with possibly a double-bass in either Eb or BBb. As 
the size of the band increases, the proportion of clarinets 

With this array of clarinets, the symphonic band has great 
resources in tonal colorings. The soprano clarinets can ascend 
into their top register and terrify with their shrill and strident 
voice. In their lower register they can blend with the top 
register of the alto clarinet to sing sweetly and dramatically 
in a great variety of emotional moods. The lower register of 
the alto blends in with the top register of the bass clarinet to 
express sonorous beauty of a solemn and sometimes weird 
character. The bottom notes of the bass are used to express 
that which is mysterious, foreboding, sepulchral. 

Today clarinets are made of grenadilla wood, of vulcanized 
rubber, and of metal, usually nickel silver. The first clarinets 
were made of boxwood, but warping of boxwood made this 
material unsuitable when the key mechanism included long 
hinges, for when the wood warped the keys would stick. 
Grenadilla wood was finally found to be the most suitable, 
and today the finest clarinets are made of this material. This 
wood is found in Mozambique, South Africa, and in the large 
island to the east called Madagascar. The grenadilla wood 
tree grows in desert areas, and it takes years to grow a tree 
large enough to be used commercially, for the growth is very 
slow. The wood, naturally, is very close-grained, hard and 
heavy. Originally dark brown with streaks of dark purple 
through it, it becomes darker with age, and after being soaked 
in oil it becomes black. The best wood is seasoned in un- 
heated wood lofts for from five to ten years and even longer. 


This aging process is carried on so the grain will become 
settled and stable. Cracking and splitting, shrinking and 
swelling are the bane of the clarinet player, and manu- 
facturers use every effort to build clarinets of wood which 
will not have these faults. 

In spite of the greatest care, some grenadilla-wood clarinets 
split, especially when used carelessly and in changeable 
climates. This has led to the use of ebonite, or vulcanized 
rubber. Other clarinets are made of metal, the first on record 
having been made by C. G. Conn in 1887 and patented by 
Mr Conn on May 23, 1889. Flutes were made of metal by 
Theobald Boehm in 1847, in Europe, but there is no record 
of a metal clarinet before 1887, either in Europe or America. 
This invention was ahead of its time, and it is only within 
the last ten years that metal clarinets have been used ex- 
tensively. They meet the same criticism as Boehm's metal 
flutes had to contend with. Some musicians claim a metal 
clarinet does not have as rich a tone quality as a wood 
clarinet. Numerous blindfold tests have been conducted in 
which musicians used both wood and metal clarinets. The 
best musicians, both players and listeners, have repeatedly 
been fooled by these tests, showing that they could not tell 
the difference between wood and metal clarinets. The preju- 
dice is strongly in favor of wood, however, and it will prob- 
ably take years before the metal clarinet is universally 
accepted. Materials do unquestionably make a difference in 
tonal quality, but few ears are keen enough to distinguish 
between wood and metal, and there are tremendous ad- 
vantages in durability, permanence of adjustment, and 
manufacturing cost in building clarinets of metal. Further- 
more, who can say that the wood-clarinet tone is more 

"beautiful" than the metal-clarinet tone? They may be 
different (although even this point is heatedly debated), but 
by what standards can we say one is more "beautiful" than 
the other? The wood clarinet may be more mellow, while 
the metal clarinet may be more brilliant, but is one, therefore, 
more "beautiful" than the other? If the argument narrows 
down to a question of whether the "true" clarinet tone can 
be produced only on a wood or on a metal clarinet, who can 
say what is the "true" clarinet tone? There are pronounced 
differences of qualities in different registers on the same 
clarinet, differences much greater than those between wood 
and metal clarinets, taken register for register. Only time and 
usage will tell, as they did in the argument about wood and 
metal flutes. Today a wood flute is a curiosity, and possibly 
the same will be true of the wood clarinets in a decade or two. 


The Saxophones 


MENTION JAZZ MUSIC and for some strange reason nine 
out of ten people think of the saxophone. Strange, 
because many other instruments are much more closely 
identified with jazz than is the saxophone. Much more of a 
jazz instrument is the cornet, trumpet, clarinet or trombone; 
and long before the saxophones entered jazz, the piano, 
violin, string bass, guitar and banjo were old-timers at it. 

The first jazz instruments were not the saxophones but the 
cornet and trumpet; in fact the germ idea of jazz music was 
originated on these two instruments. It all seems to have 
happened down in New Orleans about 1900. It was the cus- 
tom to hire bands for advertising purposes, and these bands 
would travel around the city in a band wagon, much as bands 
are used today during political campaigns. Their repertoire 
was often meager, and when once a selection had been learned 
it was given an intensive workout. In these bands the leader 
was often the cornet or trumpet player, as in the silver-cornet 
bands. Being the best musician in the band, the cornet player 
would relieve the monotony of these few stock numbers by 

adding obbligatos, according to his inspiration. These ob- 
bligatos became more and more elaborate, and playing 
obbligatos became a sort of new art at least this form of 
obbligato. As the number would be played again and again 
by the band, the cornet player would play one obbligato 
after another, each one impromptu and each one different. 

There grew up considerable rivalry among these obbligato 
artists, particularly in the colored bands. Each band thought 
it had the best obbligato player. Often these bands would 
fight against each other to see whose obbligato artist was 
the better. That cornet player who could play a chorus the 
greatest number of interesting ways, and who could play 
it the loudest and highest, was considered the best. This 
activity was called "bucking" or "carving." Improvising 
music of this kind required musical ideas and highly de- 
veloped musical technique. Somewhat as the great old mas- 
ters developed their music by stating a theme or idea and 
then restating it again and again in different forms, so these 
Negro musicians started with a melody and then played it 
over and over in endless variations. The form of the variation 
was original, however, and is the great contribution of the 
colored race to music. Whether or not one likes jazz or 
"swing" music, one has to admit that this type of perform- 
ance is something different in the history of music. There is 
a sort of wild barbarity, a spontaneity, a "swing" to this 
music which no other music has. As Duke Ellington says in 
a song of his, "It don't mean a thing, if it ain't got that 

This new idea "caught on," and it wasn't long before the 
colored dance bands were taking over the style. Not only 
did the cornet player improvise, but the other instrument 


players took their turn, or two or three would improvise to- 
gether. One of the best of these early jazz bands was known 
as King Oliver's band, and it became famous in New Orleans 
from 1915 to 1918. The instruments were at first clarinet, 
trombone, violin, string bass and drums. Later the instru- 
mentation was changed and consisted of two trumpets, 
clarinet, trombone, banjo, drums and piano. But there were 
no saxophones. 

White musicians used to drop in and listen to King Oliver 
and his band "go to town." These musicians realized they 
were hearing something new in music, and they were fasci- 
nated. They finally formed white bands and played this new 
type of music. One of the most famous of these early white 
bands was called the "Dixieland Jazz Band," organized and 
led by a trumpet player known as Nick LaRocca. This 
band came North in 1917, toured as far as New York and 
then went to England. It created a sensation, for this was the 
first time most people of the North and England had heard 
jazz music. The instruments in this band were trumpet, 
trombone, clarinet, drums and piano. But there were no 

Colored jazz bands used to play on the river boats as they 
traveled up and down the Mississippi, and some of these 
bands left the boat and played jobs in the towns along the 
river. One of these early colored bands which ventured to 
come North was the "Original Creole Orchestra." This 
organization is said to have made a trip North as early as 
1911. It consisted of cornet, trombone, clarinet, string bass, 
violin and drums. Later a guitar was added, but no saxo- 
phones were used. 

The North did not see a famous colored band until King 

Oliver brought his musicians to the Dreamland Cafe in 
Chicago in 1918. The "Dixieland Jazz Band" had met with 
such success up North that the colored boys decided they 
would try it. Their engagement was a success, for people 
liked this new kind of music. In 1921 this band made some 
recordings of their music for the phonograph, the first great 
jazz band to be recorded. The instruments in this popular 
jazz band were two cornets, trombone, clarinet, banjo, piano 
and drums. But there were no saxophones. 

It was not until the third decade of the century and of the 
history of jazz music that a saxophone was introduced into 
the jazz band. One of the first bands to use a saxophone was 
called "Joe Kayser and his Novelty Orchestra." About 1921 
this band was comprised of violin, piano, banjo, drums and 
a C-melody saxophone. Two years later we find the "New 
Orleans Rhythm Kings" playing in Chicago. The instru- 
ments used were clarinet, trumpet, trombone, string bass, 
piano, drums and a C-melody saxophone. Later a banjo was 
added. The "Wolverines" were another famous jazz outfit, 
organized in Chicago in 1923. It was comprised of cornet, 
played by Bix Beiderbecke, acknowledged the greatest of all 
jazz players; also clarinet, string bass, piano, banjo, drums 
and tenor saxophone. A year later the band was reorganized, 
and a trombone was added. 

Although there were beginning to be saxophones in the 
jazz band, this instrument was not generally accepted in the 
jazz band, as can be seen from the fact that when the great 
Bix organized his own band in 1924, he did not use a saxo- 
phone. This band, known as "Bix Rhythm Jugglers," was 
comprised of a cornet, trombone, clarinet, piano, drums and 
banjo. While it was evident that the typical jazz band re- 


C W) rt 


w m 


a) jJC; 
fl ? , 
o o <u 

iL / A ; * JL* ^^^^^.>m^ rt .^-L.:^&^M<i^^^^m^mi^^i^ 

cr , 

rt 5" 
<U ^ 



(1) Military serpent used in Europe first part of the eighteenth century. 

(2) Ophicleide. Instruments of this type were used at the Battle of Waterloo 
in 1815. (3) Original Adolphe Sax saxophone, invented by Sax in 1840. (4) First 
saxophone built in America, by C. G. Conn in 1887. (5) Modern alto saxophone 
in Eb- 


quired cornet, trumpet, trombone, clarinet, drums and piano, 
with string bass, guitar, violin and banjo often called for, the 
saxophone was looked upon as a sort of novelty instrument. 
Whenever a saxophone was called for it would usually be 
the C-melody. Not until considerably later were the alto, 
tenor and baritone saxophones found in the jazz band. 

That the saxophone was not used more in the jazz band 
is all the more remarkable when it is considered that there 
was a saxophone craze in America from 1919 to 1925, reach- 
ing a peak in 1923 and 1924. In these two years it is estimated 
there were one hundred thousand saxophones made and sold 
in America annually, besides those imported. During those 
six or seven years of the craze, the total saxophones sold in 
America ran well over a half-million instruments. Yet, in 
spite of this great popularity among amateurs, there were 
astonishingly few saxophones used in the jazz band. This 
fact, as much as any other, would stamp the saxophone as 
anything but a typical jazz instrument. 

Selling the saxophone to the American public has often 
been credited to Tom Brown and his famous saxophone 
sextet. He it was who taught America that the saxophone 
can moan, laugh, cackle, titter, squeal and grunt. To many 
uninitiated, this perversion of the saxophone talents is 
synonymous with jazz, although it really has nothing to do 
with it. There is good reason to believe that much of the 
general misconception about the saxophone's being a jazz 
instrument can be traced to this wrong notion. Tom Brown 
organized his saxophone sextet in 1911 and continued it 
until 1926, reaching a peak of popularity about 1924, which 
through more than a coincidence was the peak of the saxo- 
phone craze. After this date, both Tom Brown and the 

saxophone slumped in popularity, which shows how the 
popularities of both were interlocked, 

Tom Brown's saxophone sextet had many imitators. The 
vaudeville stages from 1915 to 1925 were crowded with 
them. They cavorted around the stage with a successful 
combination of antics and music which pleased the audience. 
Much of the attraction of the saxophone for early audiences 
before which Tom Brown performed was the comparative 
novelty of the instrument. Outside of a few concert bands 
the saxophone was hardly known. But Tom Brown was a 
great showman, and he made the saxophone the most-talked- 
of instrument in America. After a few years there were scores 
of acts like his, and within a few years thousands of people 
began buying the C-melody saxophone because they yearned 
to play this amazing instrument. They also bought altos and 
tenors and sopranos and baritones and formed amateur 
saxophone ensembles and saxophone bands. And although 
the American people seemed clear that their own amateur 
playing was not jazz, they seem never to have properly 
distinguished between Tom Brown's antics on the saxophone 
and the antics of the real jazz player. 

Many lovers of the saxophone now wish the instrument 
had been introduced to the American public through the 
concert band rather than through Tom Brown and his 
imitators. It no doubt would have taken much longer, but 
the reputation of the saxophone would have been much more 
favorable. E. A. Lefebre played the saxophone for Gilmore's 
band with great virtuosity. Other great legitimate artists 
were Jean Moeremans, soloist with Sousa, and BenVereecken, 
soloist with Pryor. H. Benne Henton, saxophone soloist 
with Sousa, showed thousands that the saxophone could be 



played with artistry as great as that expected in singing or 
playing the violin. It was he who was chosen by Richard 
Strauss to head the saxophone quartet which played the 
Strauss "Domestic Symphony" when the great German 
composer toured America. 

The almost fatal reaction from the so-called saxophone 
craze, when C-melody saxophones squawked in every block, 
was that anyone could pick up the saxophone and play it 
after a few hours of practice. The saxophone is surprisingly 
easy to play in a simple way, but although the rudiments 
can be picked up quickly, it takes years of diligent study 
under competent teachers to master the instrument and 
play it with artistry. Only recently, more than ten years 
after the craze, has this sane view of the saxophone become 
at all general. The saxophone is finally taking its place as a 
serious instrument alongside the trumpet, trombone, violin 
or cello. 

By 1926 it was customary for the regular dance band to 
use two or three saxophones, but dance bands are by no 
means all jazz or swing bands. The tendency for the dance 
band is to grow larger and larger, but the typical jazz band 
remains small; and while today there are some fine swing 
bands with ten to a dozen players, the true "jam" band 
the jazz or swing band of the purest type seldom numbers 
over six or seven. In such bands there may or may not be 
a saxophone, but there usually are a cornet or trumpet, a 
trombone, a clarinet, a string bass or drums and a piano. 

There is no more reason for dubbing the saxophone a 
"jazz instrument" than there is for applying the same term 
to the violin. Did anyone ever play purer swing music on 
the saxophone than Joe Venuti plays on the violin? Why pick 

on the saxophone when the clarinet figures so prominently 
in jazz? Did anyone ever "swing" better on the saxophone 
than Benny Goodman and Peewee Russel do on the clarinet? 
And why hold the saxophone up to ridicule and say nothing 
of the trombone? Did ever a saxophonist get "hotter" on 
his instrument than Jack Teagarden and Tommy Dorsey 
do on trombone? And what about Louis Armstrong and 
Bunny Berigan on cornet? Earl Hines and Teddy Wilson 
on piano? Gene Krupa on drums? Vic Berton on tympani? 

On the other hand, who is prepared to say that Lucien 
Cailliet, first saxophone of the Philadelphia Orchestra, 
Holland Tapley, first saxophone of the Boston Symphony, 
Leonard Schaller, first saxophone of the Chicago Symphony, 
Carroll Gillette, first saxophone of the San Francisco Sym- 
phony, or Maurice DeCruck, formerly first saxophone of the 
New York Philharmonic-Symphony, is less of an artist than 
flutists, violinists or cellists are on their instruments? Who- 
ever has heard any of these great artists play the troubadour 
song at the castle, in "Pictures from an Exhibition/' by 
Moussorgsky-Ravel, is willing to class the saxophone among 
the most beautiful of musical instruments. Other convincing 
proof is the saxophone solos in Ravel's "Bolero" and De- 
bussy's "Rapsodie" for saxophone, as played by any one 
of these fine saxophone artists. It is not the instrument at 
all, but the way it is played, which makes it legitimate or 

Practically every instrument has been used in jazz, and 
the only reason all of them are not used is that many of them 
do not have what it takes to play jazz. One of the character- 
istics of jazz or swing music is a most vigorous attack. The 
cornet, trumpet, trombone, clarinet, saxophone, drums, piano 


and string bass all have the "sock" necessary for this 
attack, but how could a player "wham 'em out" on the 
flute, oboe, English horn, bassoon, French horn, viola or 
harp ? Even the violin, though often used, is not a good jazz 
instrument because it lacks the punch necessary. If it is 
true that the saxophone got into bad company, the only 
reason the other instruments escaped is that they were too 
weak and timid. 

When Adolphe Sax invented the saxophones in 1840, he 
expected them to take their place in the symphony orchestra 
and concert band along with other legitimate instruments. 
In France particularly, and also to less extent in Germany, 
the saxophone was readily adopted and was given important 
parts in opera and symphony music. In France, Meyerbeer, 
Massenet, Berlioz, Bizet and Saint-Saens have used the 
saxophone to good effect, while in Germany Richard Strauss 
has been a champion of the saxophone, using a quartet of 
them in his "Domestic Symphony" and other compositions. 
Modern composers in Russia, Italy and America are giving 
more thought to the use of the saxophone resources, and 
considerable good music is being composed for this worthy 

In the concert band, too, the saxophone is earning a solid 
place for itself. Sousa always used from four to eight saxo- 
phones and said one time, "There is much to be done in 
standardizing the saxophone, with its strange sweetness of 
tone and its variety of effects." Composers and arrangers 
are beginning to recognize that the saxophone family holds 
a unique place in the tonal spectrum, between the woodwinds 
and the brasses; for the saxophone is a strange blending of 
these two colorings, partly woodwind and partly brass. They 

serve a most useful purpose of blending the woodwind choir 
and the brass choir together. 

The invention of the saxophone, by Antoine (commonly 
known as Adolphe) Sax in 1840, was an accident. Sax was 
conducting some experiments with the ophicleide, an old 
type of cup-mouthpiece instrument then in use but now 
obsolete. This instrument is more completely described in 
the chapter on the tubas. It was made of brass, the holes 
were closed by keys much like today's saxophone keys, 
and it was played with a cup mouthpiece. Sax was curious 
to learn what would happen if he played the instrument 
with a clarinet mouthpiece. 

The strange blending of the brass and reed tone fascinated 
Sax. It was somewhat brassy, from the brass body, but it 
did not sound like the old ophicleide. It was reedy, from the 
clarinet reed mouthpiece, but it did not sound like a clarinet. 
It did not behave like a clarinet, either, for while the clarinet 
has only the odd partials in the scale, this new cross between 
a brass instrument and a clarinet had both the even and the 
odd partials. Sax realized he had produced a new tonal 
coloring among musical instruments, and after further ex- 
periment, during which he gave the saxophone body inside 
dimensions different from those of the ophicleide, he pro- 
duced the instrument known today as the saxophone. 

This crossbreeding of the clarinet with the ophicleide 
produced an instrument which not only had both even and 
odd partials, but which sounded a fundamental, or lowest, 
note whose wave was only twice as long as the instrument. 
In the chapter, "How Music Is Made," it is pointed out 
that the clarinet can sound a low note whose wave is four 
times the length of the instrument. This latter difference 


is not as important as the former, for in the saxophone, 
which has both even and odd partials, the key difficulties 
found on the clarinet are eliminated. While the distance 
between the first and second partials, or open notes, is a 
twelfth on the clarinet, it is only an octave on the new 
hybrid saxophone. The complicated mechanism necessary 
to bridge the gap between these two widely spread notes 
on the clarinet is not necessary on the saxophone. The 
fingering for the first octave can be repeated in the second 
octave by simply overblowing to the octave above, the same 
as on the flute and oboe. This enables the maker to build a 
key mechanism which is less complicated than the mecha- 
nism on the clarinet. It is this comparative simplicity of 
the key mechanism which partly accounts for the great 
popularity of the saxophone during the saxophone craze. 

It is one of the mysteries of instrument making why the 
clarinet mouthpiece on the clarinet produces an instrument 
with the characteristics of the closed pipe, while the same 
mouthpiece on the saxophone produces an instrument with 
the characteristics of the open pipe. Why should the clarinet 
have only the odd partials and a fundamental whose wave is 
four times the length of the instrument, while the saxophone 
has both even and odd partials and sounds a fundamental 
whose wave is only twice the length of the instrument? 
This can be explained only in rather technical terms, which 
are out of place here. But some hint of the principles involved 
can be given here by pointing out that the body of the 
clarinet is cylindrical while the body of the saxophone is 
conical. Delicate acoustical apparatus has shown us that the 
even partials are actually present on the clarinet, but they 
are so weak they cannot be used. It seems that the cylindrical 

tube of the clarinet in some way dampens out the even 
partials. Some of these upper partials are quite strong, such 
as the eighth and tenth. In fact, these two even partials are 
of practically the same strength as the odd ninth. One theory 
advanced is that the natural frequency of the reed is approxi- 
mately the frequency of the eighth and tenth partials and 
that the reed frequency strengthens these two and makes 
them stronger than the other even partials. The natural 
frequency of the reed is that sounded when it is struck or 
when it is blown on the mouthpiece unattached to the in- 
strument. Under both these conditions, the pitch is high 
and lies near these two even partials. However, the partials 
which could be useful, such as the second and fourth, are 
so weak they can scarcely be detected by the ear. On the 
other hand, the conical tube of the saxophone shows no 
preference for even or odd partials and produces both with 
equal strength. 

The mouthpiece is also an important factor to be con- 
sidered in comparing these two instruments. The flute, which 
also has a cylindrical body the same as the clarinet but a 
different mouthpiece, produces both even and odd partials. 
If a clarinet mouthpiece is inserted on the flute, the even 
partials drop out and the instrument takes on all the charac- 
teristics of the clarinet. The single-beating reed of the clarinet 
closes the end of the instrument and makes it a closed pipe. 
The end of course is not closed all the time, for the reed opens 
and closes alternately. But on the clarinet, with its cylindri- 
cal tube, the single-beating reed acts as if it closed the tube. 
The reed vibrates in a similar manner on the saxophone, 
with its conical tube, but on this instrument the effect is 
not that of closing the tube. This is explained by acoustical 


engineers by the degree in which the reed closes the tube of 
each. The reed on a clarinet, vibrating at the same frequency 
as it does on the saxophone, seems nevertheless to render 
the tubing closed more of the time than it is open, and gives 
the clarinet the characteristics of the closed pipe. 

We speak of the clarinet mouthpiece and of the saxophone 
mouthpiece as if they were identical. They are not identical, 
but only similar. They are alike in having a single-beating 
reed which opens and closes a chamber in the mouthpiece. 
The chamber in each, however, is slightly different in shape 
and size. These small differences have been developed as 
refinements in each mouthpiece to favor the scale and re- 
sponse of each instrument, but both mouthpieces remain 
alike in principle. 

The inventor built a large family of these instruments, 
including, according to Berlioz, the following: the high Eb 
soprano, the Bb soprano, the Eb alto, the Bb tenor, the 
Eb baritone and the Bb bass. Other members of this prolific 
family today are the C soprano, the C tenor, more familiarly 
known as the C-melody; and the F alto, also called the 
F mezzo-soprano. There are also a few contrabass saxo- 
phones in Eb, but these are rare. Add to these distinctly 
different members of the family some different shapes for 
some of them and you have a large family. Among these 
varieties with different shapes are the straight and curved 
Bb sopranos, the straight and curved F mezzo-sopranos, the 
curved Eb alto and the Eb alto which is mostly straight but 
turns up at the end something like a clay pipe or hockey 

Although made in this great variety of keys and shapes, 
the most useful members of the saxophone family are the 

middle voices: the Eb alto, the Bb tenor and the Eb baritone. 
Next in importance are the Bb soprano, the F mezzo-soprano, 
and the Bb bass. The other members of the family are not 
commonly found in bands and orchestras. Occasionally the 
little Eb soprano is used for special oriental effects, and some 
jazz musicians prefer the C-melody, but they are heard 

It could hardly be expected that this new family would 
become popular overnight, but gradually the composers 
came to appreciate these instruments and give them parts 
in both orchestra and band. In America there was little 
known of the saxophone from 1840 to 1880. Only a few were 
imported from Europe, and these were practically useless 
from lack of music. About 1850 C. G. Conn began to make 
a few saxophones in America. His cause was greatly helped 
when E. A. Lefebre, great saxophone virtuoso of Europe and 
personal friend of Sax, came to America and toured the 
country as soloist with Patrick Gilmore's celebrated band. 
In 1895 Lefebre was employed by Mr Conn to supervise the 
manufacture of the American saxophone. Thirty years later, 
as previously related, America was buying more saxophones 
in a week than Antoine Sax ever dreamed would be sold 
anywhere in a year. 

It was this tremendous demand for saxophones which 
brought about a unique condition in their manufacture. 
The American genius for quantity production was turned 
to the manufacture of the saxophone. In former days the 
saxophone had been made slowly by hand, piece by piece. 
When the great demand for saxophones hit the country, 
modern methods had to be employed. Precision machines 
were developed, new processes were invented and new 


accuracy and speed obtained. Today, although the demand 
has lessened, saxophones are still the largest-selling family 
of wind instruments. These modern manufacturing methods 
have been continued and refined, and the manufacture of 
the modern saxophone in America is now an interesting 
process, which amazes those who see it for the first time. 
Few people who play or listen to a saxophone realize it is 
made of over five hundred separate and individual parts, 
efficiently and accurately assembled into a fine musical 
instrument. Among these parts are over five dozen small 
key castings, four dozen knobs for holding the key hinges, 
four dozen hollow hinges of various lengths, four dozen 
springs and pads of assorted sizes and kinds, and over five 
dozen screws, from long ones nearly ten inches in length 
to tiny ones weighing 1,240 to the ounce! 

The five and a quarter pounds of raw material in a high- 
grade saxophone are not worth over five dollars, but the 
finished instrument retails to the musician for over one 
hundred dollars. The greatest cost which goes into a saxo- 
phone is that of the labor of skilled and long-experienced 
craftsmen. For example, there are thirty-three keys on the 
common alto saxophone, and on each key there are, on the 
average, thirty-four separate and distinct operations, such 
as casting, polishing, milling, drilling, brazing, locating spring 
hooks, buffing making a total of over noo separate opera- 
tions to produce a single set of keys for one alto saxophone. 
Nor can a saxophone be cut out and sewed together like a 
suit of clothes or be kneaded by the hands into shape like a 
loaf of bread. Saxophones are made of brass, and hardly 
any operation can be performed without some expensive 
lathe, drill press, screw machine, punch press or other special 

machine, for brass cannot be worked by hand but only by 
accurate machines which can shape it into a finished product. 
The cost of these tools and machines runs into thousands of 

The saxophone today is substantially the same as it was 
when Sax invented it nearly a hundred years ago. Of course 
it is more efficient and more accurate, but in appearance 
and general features it has changed but little. At first the 
saxophone was built only down to low B, but now nearly 
all instruments go down to low Bi?. At first the top note 
was D; now nearly all members of the family can play up 
to F, The first saxophones were provided with two octave 
keys; now nearly all saxophones have the single automatic 
octave key. On the modern saxophone, also, are many im- 
provements in mechanism which make difficult passages 
easier, especially rapid trills. 

If Antoine Sax could see and hear the modern saxophone 
played today by a "hot" jazz artist, he would look at it and 
say, "That's my invention, all right, but I never dreamed it 
could do things like that," If he could see and hear it played 
by a great soloist in the symphony or concert band, he 
would say, "That unquestionably is my saxophone, but I 
never hoped it could sound so beautiful." And if he could 
see it being made in a modern American factory, he would 
say, "I never thought of making them that way. But tell 
me what on earth do you expect to do with them all?" 


The Trumpet and the Cornet 


MOST LAYMEN don't know the difference between a 
trumpet and a cornet, but among musicians the 
trumpet versus the cornet is a subject which calls for heated 
argument. For centuries there has existed a bitter feud be- 
tween these two instruments. First one would rise to the 
ascendancy and then the other. When one would be rising 
in favor, the other would be declining. Before long the tables 
would be reversed and the second choice would find itself 
swept upward on a wave of acclaim, while the other would 
decline in popularity. This seesaw feud has gone on for hun- 
dreds of years and is no nearer solution now than it has 
ever been. 

As is true in most feuds that are especially bitter, this is a 
blood feud. The closer related by blood certain families or 
tribes may be, the more bitter the feud, and this is true of 
the contest between the trumpet and the cornet. When 
closely related in this manner, the only way to carry on a 
heated feud over a long period of years is to keep the fancied 
issue clearly defined; otherwise common blood and common 

interest will cause the hatred to cool off, and the feud will 
die a natural death. In order to understand the feud between 
the trumpet and cornet we must first understand why two 
instruments so similar can find enough difference between 
themselves to make a good fight. 

The trumpet is the blue-blood in this feud. It points with 

1000 1100 120O BOO I40O l?OO I60O 170O I6OO 1900 2000 

TRUMPETS (Straight Bore) 

CORN ET5 (Tapered Bore) 

Trumpets with their straight bore held the ascendancy for 
centuries over the cornets with their tapered bore. Late in the 
thirteenth century the rival cornets appear in churches and in 
early opera and reach great popularity in the sixteenth. At this 
time the trumpets languish. By 1750 the cornets give place to the 
oboes, and the trumpets steal a march in the roles given them by 
Bach and Handel. In 1800 trumpets wane when clarion parts are 
given to clarinets. The keyed bugle takes a sharp rise right after 
1800, a triumph for the tapered cornet, but before it could rise 
high, the piston valves put the trumpet into the ascendancy again. 
During the era of the cornet virtuosi, just before and after 1900, 
the cornet was the favorite, but as the silver-cornet band gave way 
to the more modern concert band, the trumpet arose to claim 
popular favor. Now it looks as if the cornet were staging a come- 

great hauteur to its ancient ancestors among the Chinese, 
Egyptians, Greeks and Romans. The genealogy of the trum- 
pet can be traced in the records of the kings and princes, 
because for centuries the trumpet was the badge of royalty 
and could be used only by royalty or in royalty's service. 
Even as late as the Stuarts of England, the unauthorized 
possession of a trumpet was a serious offense. Present at all 
court functions, celebrations of great martial victories and 

important affairs of state, the family tree of the trumpet 
grew side by side with the family tree of royalty. 

The lineage of the trumpet can be traced through its 
cylindrical bore. From two thirds to three fourths of its 
length was of uniform diameter with the opening at the 
mouthpiece end. At the opposite end was a length of tubing 
of increasingly larger bore, ending abruptly in a bell. The 
tone of the trumpet was brilliant and martial. It spoke in 
a commanding and confident manner, as the king spoke. 

During these early centuries there undoubtedly existed 
another instrument similar to the trumpet but having a 
conical, or tapered, bore. Little has been written about such 
an instrument, as would be expected in the case of such a 
commoner. This instrument was much like our common 
bugle in bore and tone quality. It was small at the mouth- 
piece end and had a wide angle taper in the bore which 
flared out at the opposite end in a bell. Its tone was round 
and mellow, better suited to singing than to command- 

Originally both the aristocratic trumpet and the plebeian 
bugle were a single, straight tube. Both were played with a 
cup-shaped mouthpiece. There was little in outward appear- 
ance to distinguish one from the other. Later it became the 
custom to bend the trumpet in the shape of the letter S. 
The bugle-type instrument was quick to follow this custom, 
and again the aristocrat and the commoner were difficult 
to distinguish apart. Still later the trumpet was bent in a 
sort of flat loop, like that characteristic of the ordinary 
bugle of today, and the commoner was not far behind in 
adopting this shape. Outwardly these two instruments were 
similar. Their musical natures, however, were different: the 

aristocratic trumpet was cylindrical in bore, the plebeian 
bugle-type instrument was tapered, or conical, in bore. 
This was the issue between them which has kept alive the 
blood feud for sixty centuries or more. 

Even as the army lords, kings and priests held undisputed 
sway over the common people for many centuries, so the 
trumpet held sway over the bugle-type instrument. But 
along in the Middle Ages something happened both to the 
common people and the plebeian bugle. Just about the time 
the English barons struck for the common people against 
the oppression of kings and forced King John to sign the 
Magna Charta at Runnymede, the bugle-type instrument 
overthrew the oppression of the trumpet through a kind of 
intermarriage with royalty. The pure blood stream of the 
trumpet became contaminated with the blood of the common 
bugle, and the bastard cornet was born. And woe has been 
the lot of the trumpet ever since. This crossbreeding seems 
to have produced a surprisingly virile offspring. The old 
bugle-type instrument had not been able to contend at all 
successfully with the blue-blood trumpet, but the half-breed 
cornet seemed to incorporate some of the virtues of each 
which gave it a better chance in the long feud with the 
trumpet. The cornet adopted a bore which was partly 
straight and partly tapered. Its tone was not as round and 
mellow as that of the old bugle-type instrument, nor yet as 
brilliant and piercing as that of the trumpet. It could sing 
like its plebeian ancestor, although not with such a golden, 
mellow voice; and it could speak like its royal ancestor, al- 
though its voice lacked some of the silvery brilliance and 

By the fifteenth century the trumpet had been crowded 

into second place for the first time in history. The pure 
tapered-bore instrument had never been able to do it, but 
the bastard-bore cornet had been more successful. By the 
sixteenth century the cornet had all but put the trumpet 
into obscurity. Everywhere the cornet flourished in the 
choirs and churches and in the first orchestras started in 
Italy by the creators of opera. This continued until the 
beginning of the eighteenth century, when a double blow all 
but knocked out the cornet. The oboe became suddenly a 
very popular wind instrument and usurped the place of the 
cornet. About the same time or a little later, high-pitched and 
florid trumpet parts became popular, reaching their greatest 
vogue during the time of Bach and Handel. While the oboe 
ran the interference and blocked the cornet, the trumpet 
raced on to score a touchdown, to use a football metaphor. 
- But by the end of the century there was a reaction against 
these shrill, piercing trumpet fireworks, and clarinets were 
used instead of trumpets, notably in the music of Mozart. 
While the trumpet was staggering from this blow, the cornet 
took a new lease on life when the keyed bugle was invented 
in the early nineteenth century. Many musicians acclaimed 
this new instrument as destined to transform band and or- 
chestra music; but its popularity was short, for the piston 
valve was invented a few years later, and both types of in- 
struments shared in this great improvement. In less than a 
hundred years, though, the cornet had again climbed to 
first place in popularity, through the great virtuosity of 
such well-known cornet soloists as Levy, Liberati, Hoch 
and Bellstedt. The pendulum then swung quickly to trum- 
pets, and cornets have had to take a back seat for the last 
twenty-five years. At the present time, however, there are 

many indications that the cornet is about to stage one of its 
celebrated comebacks. 

And what happened to the old, original tapered-bore 
instrument? Did it get completely put out of the running 
by the tough competition of its ancient rival, the trumpet, 
and its own bastard offspring, the cornet? No; trailing com- 
fortably behind the two more popular instruments, the 
tapered-bore instrument adopted the piston valves and has 
finally settled down to a modest but important job in the 
concert band as the fluegelhorn. 

This synopsis of the blood feud between the royal trumpet 
on the one hand and the plebeian bugle and the bastard 
cornet on the other gives an idea of the rocky road over which 
have come the two sopranos of today's brass choir, to say 
nothing of the mezzo-soprano fluegelhorn. But such a synop- 
sis necessarily omits the drama and thrills of this desperate 
struggle among themselves and the common battle they 
waged against other instruments along the way. 

The beginning of the story of the trumpet and cornet is 
something like the beginning of the story of the human race. 
The best authorities now agree that man did not descend 
from a monkey but from some primate which was the an- 
cestor of both man and the monkey. Likewise, we are able 
to trace the history of the trumpet to a common ancestor 
from which descended not only the trumpet but also the 
cornet, horn and tuba. The animal horn or the elephant tusk, 
hollowed out and cupped at the small end, may have been 
this common ancestor; or it may have been a giant shell of 
some sea nautilus, or merely a hollowed-out piece of wood. 
The only essentials of such an ancestor are a tube with a 
cup-shaped mouthpiece at the small end and a flare or bell 

at the opposite end. The mouthpiece served as a fixture 
across which the lips of the player could be stretched, and 
the tube and bell served to amplify and give resonance to 
the weak, thin buzzing of the lips. 

One of the first things early man devised after rising above 
the Stone Age into the age of metals was a metal trumpet. 
Ancient ruins reveal trumpets made of bronze, iron or other 
metals. At first these instruments resembled the animal- 
horn and the elephant-tusk prototypes which our ancestors 
copied in metal. Gradually they became more slender and 
longer as the primitive craftsmen became more familiar 
with the metal medium. From a foot long they grew to be 
three and four feet long, sometimes longer. Convenience in 
carrying and comfort in holding limited the length of these 
primitive trumpets, but we can imagine that the pitch of 
the note also had something to do with it. The fundamental, 
or lowest, note of a four-foot trumpet was approximately 
an octave below Middle C; and the next note, probably the 
"pitch" note, was Middle C, somewhere near what would 
be desired when used in the temple or on the battlefield. 
Such a trumpet had a range strikingly similar to the range 
of a male voice. A tube twice as long would have as its fun- 
damental, or lowest, note the C two octaves below Middle 
C, which was too low to be of practical use. The second note, 
probably the "pitch" note, would be one octave below 
Middle C, also too low for usefulness. Of course, higher notes 
could be sounded on these longer tubes, but we can imagine 
that the second open note was the easiest to blow on such 
crude instruments (just as it is on the ordinary bugle of 
today) and that this was the note most often used. On the 
other hand, difficulty of blowing high notes on shorter 

trumpets probably restricted the shortness of the tubing. 
For instance, a trumpet only a foot long would sound as its 
fundamental note the C which is an octave above Middle C; 
and its "pitch" note might be an octave higher, due to the 
smallness of the instrument and the difficulty of playing the 
fundamental To blow a note of such high pitch would be 
difficult, especially on an instrument as crude and inefficient 
as this was. 

So we see that physical conditions convenience in carry- 
ing and holding, and limitations of the player's lips and 
lungs naturally restricted the length of the early trumpet 
to about four feet. It is more than a coincidence that trum- 
pets and cornets of today are about that length. The Bb 
trumpet is the most used today, and -the length of its tubing 
is just about four feet. The C and D trumpets are sometimes 
played, and occasionally the Eb trumpet is called for, but 
these shorter trumpets are hard to play, as any trumpet or 
cornet player can testify. Of course, cup-mouthpiece instru- 
ments are made much longer than four feet these days, but 
this has come to pass after centuries of experimenting with 
proportioning the bore and refining the workmanship so that 
it would be possible for the player to make these longer tubes 
sound more easily. These instruments with longer tubing are 
used for musical purposes and are not restricted to uses in 
the temple or on the battlefield, as were the early trumpets. 
Low, sonorous notes are desirable for effects and colorings 
in music but were not suitable for signal purposes and for 
exorcising evil spirits. 

In China and India two or three thousand years B.C., we 
find trumpets being used to scare away evil spirits. These 
instruments had to be loud in order to be useful They were 

usually over a foot long and less than four feet. Trumpets 
are mentioned often in the Bible. The Hebrews possibly 
learned about the trumpet while they were captives in 
Egypt, but it became rapidly popular, if we are to believe 
the Hebrew historian Josephus. He wrote that during Solo- 
mon's time there were two hundred thousand in use. Such 
trumpets were about twenty-one inches long, and some were 
made of precious metals. They were used mostly in religious 
worship but were also employed on the field of battle. The 
classic instance of this is the taking of Jericho, whose walls 
are said to have crumbled while trumpeters marched and 
played around the city. 

The Greeks have a historical example of the use of the 
trumpet which is equally famous. During the siege of Troy 
the Greek trumpet, called the salpinx, shared honors with 
the spear of Achilles and the chariot of Agamemnon in taking 
this ancient city and rescuing the beautiful Helen. The 
Greeks had six varieties of trumpets and horns, but the sal- 
pinx was the best known. Trumpet playing became a favorite 
art with the Greeks, and Timoeus and Crates, two trumpet 
players, have made their names immortal by winning the 
musical contest at the Olympian games in 396 B.C. The 
Romans had a small-bore trumpet called the lituus. It was 
a straight, cylindrical tube for most of its length, but near 
the bell end it flared out at right angles like a hockey stick. 
In 1827 excavators of a tomb at Cerveterie discovered one of 
these ancient trumpets, and it is now preserved at the 

When the Roman legions conquered Gaul they showed the 
natives some of the finest metal trumpets these barbarians 
had ever seen. Other Europeans who missed this musical 

treat later trekked down to Rome and saw for themselves. 
When they sacked this great city they carried back with 
them some fine specimens of trumpets. During Chaucer's 
time the trumpet was familiar to all, as is seen from the 
many times he mentions it. In the Canterbury Tales he has 
the knight talk about the "trompes" and "clariounes." 
In the same collection of stories he has the squire refer to 
the "trumpette" and the "claryon clere." 

Out of the maze of names used to refer to these instruments 
we conclude that there were two kinds of trumpets. There 
was the large trumpet, or buzine also called trombe, 
tromp, trompe which later became the sackbut, or trom- 
bone. It was a straight tube about six feet long, to which a 
slide was later added. The other variety of trumpet was the 
clarion, an instrument shorter and of smaller bore. The 
words we use today to refer to these instruments are much 
more definitive. One we call the trumpet, which was formerly 
spelled "trurnpette" and means, in Italian (trombetta), "little 
trumpet"; the other we call trombone, which means, in 
Italian, big trumpet. The trumpet and trombone are so 
similar in bore and tonal quality that they really are little 
and big brother of the same family. 

During or a little before the fourteenth century, while 
Chaucer was making immortal in his writings the names 
trompe and clarioun, a new instrument put in an appearance 
in Europe which was destined to challenge the supremacy of 
these aristocratic instruments. In Germany it was called 
the fcinke, and in England it was called the cornetto. If ever 
an instrument belonged to the lower strata of society, the 
cornetto did. It was originally made of wood and covered 
with leather. It had six finger holes in front and a thumb 


hole behind and was played with a cup-shaped mouthpiece 
made of wood. The tone was dull and windy, but for some 
strange reason it became extremely popular. In the fifteenth 
and sixteenth centuries it became the most popular wind 
instrument in Europe. It was employed, along with the 
sackbut, to play with the soprano voices in the church choirs. 
In the old account books of the churches of the time there 
are many records of payments to this or that cornetto player 
for his services. When the music-loving King Henry VIII 
died in 1547, he left a noble collection of twenty of these 
odd instruments. 

Possibly one reason for the popularity of the cornetto 
was the ease and cheapness with which it could be made. 
Anyone with a pocketknife, a stick of wood, a piece of thin 
leather and a gluepot could make one. Two pieces of wood 
were hollowed out and glued together to make a conically 
shaped horn. This crude wooden horn was then covered with 
leather, and holes were drilled into the side. The mouthpiece 
was equally simple to make, being similar in shape to today's 
mouthpiece for the cornet, only it was made of wood. There 
were three sizes in use in England, the smallest being the 
treble cornetto in F, about eighteen inches long. The next in 
size was the cornetto in C, about two feet long. The largest 
was the great cornetto in G, nearly three feet long. In Ger- 
many these instruments were made in other keys, the small 
treble zinke in D being a common one. This member of the 
family was just about a foot long and was unusually difficult 
to blow and poor in tone quality. 

^When opera began to develop in Italy, toward the close 
<jf the sixteenth century, it was only natural for the com- 
posers to look to the cornetto as an instrument for the 


orchestra. The chief form of music before opera sprang up 
was church music, and the cornetto was one of the star 
performers in the choir. It is not surprising that in 1565 
Striggio and Corteccia used four cornettos in their odd 
assortment of about two dozen instruments, used to accom- 
pany light plays. This newcomer among musical instruments, 
with a background of less than three hundred years, had 
crowded out the celebrated trumpet, which had lived with 
kings and princes for several thousands of years. No trumpets 
were allowed in this motley group of lutes, flutes, flageolets 
and violins, although four trombones were admitted. * 

In 1600 Jacopo Peri composed "Euridice," the first opera 
ever performed in public. In the same year Emilio del Cava- 
liere produced his previously referred to oratorio, "La 
Rappresentazione delP Anima e Corpo." Although neither 
the cornetto nor the trumpet was called upon to perform in 
the orchestras which provided instrumental music for these 
productions, this was more of a reflection on the trumpet 
than upon the lowly cornetto. Eight years later, however, 
Monte verde called for two cornettos, one small clarion and 
three trumpets, as well as four trombones. This was quite a 
"brass" section, even though there were about twenty 
stringed instruments in the orchestra. The interesting part 
about the use of trumpets was that Monteverde called for 
them to be played with mutes. This is thought to be the 
first time mutes were used. After Monteverde, Legrenzi of 
Venice wrote for two cornettos, while Lully, the French 
composer, preferred the trumpet, using it for volume and 
color effects. Bach used the cornetto sometimes but was more 
familiar with the trumpet, usually writing for three of them. 
Handel, Bach's contemporary, apparently did not think so 


i-i 4J 

o o rt 

S 2 <y bfi 


3 CV~ r 
S c 

PJH a; S 

CT3 55 3 

w a) u, 

P . * r- V 
*-i p W 

o cuh c 
^ - 


S fl S 
G rt Ji 

CD ^ ^ ^T >> 

C rt j^ 

8 ^ | 

** o ^ ^ ^ 

QH ^ ^ 5 n 

H o <u ^ 

* S ^-S S 

.p ^ -P 

c c 

.S C 


^ 4J OJ 

C- 5 * 
*S S3 

o ^ 

j^H ^ 

w t C 
v i-n o 



.S S 


H O T3 M c 
B c <u .. e 
fa fc ij 




M +J C 

highly of cornettos, for it is said he experimented once with 
them and that was enough. 

The cornetto declined toward the close of the seventeenth 
century. In England it was supplanted by the oboe, and by 
the middle of the eighteenth century few were left. In Europe 
it survived until Bach and Handel. Gluck was one of the 
last composers to use this type of. instrument, the zinke 
being found in some of his later scores. Few cornetto or 
zinke parts are to be found in any scores later than 1780 to 
1795. Although the soprano of this type of instrument be- 
came obsolete at this time, the bass member of the family, 
the serpent, or ophicleide, survived until well into the nine- 
teenth century, no doubt because there was no other and 
better wind bass to take its place. 

As the cornetto declined, the trumpet increased in popu- 
larity, until it reached a peak during the time of Bach and 
Handel. Monteverde at the beginning of the seventeenth 
century started the style of scoring for one small trumpet, 
or clarion, and three regular trumpets. Later it became the 
custom to write for first and second clarions in the upper 
parts and for trumpets in the lower parts. Stradella and 
Pallavicino, two Italian composers, further developed the 
vogue of writing specially high parts for the clarion, and by 
1700 Scarlatti, Pufcell and others had established the manner 
of featuring the clarions in extremely high and ornamental 
solo passages. This custom reached its peak during the time 
of Bach and Handel, but after their time this practice de- 
clined rapidly. 

This era is of especial interest to the trumpeter. Many of 
the trumpet, or clarion, parts written during these years 
are almost impossible to play today on the regular trumpet. 

This has caused much conjecture about how such music was 
originally performed, if it were really performed at all, on 
the trumpet. To meet the demands of the composers in 
those days certain players, fitted to the task, became clarion 
players, while others became trumpet players. The typical 
clarion was a short instrument of small bore and was played 
with a small, shallow mouthpiece. Clarion players developed 
a freak lip, practising only in the very top of the scale and 
attaining a proficiency and skill especially for playing these 
extremely high notes. Trumpet players today know that 
they can increase their range upward by special practice 
and by using small and shallow mouthpieces, but by special- 
izing on these high notes and by using a small, shallow 
mouthpiece they lose the ability to play well in the lower 
part of the trumpet scale. 

In the time of Bach and Handel, the clarion player was 
required to play only in the top of the scale. He was a special- 
ist. Lower parts were given to regular trumpet players. This 
explains why these high parts today are so difficult, if not 
impossible, for the regular trumpet player. No one attempts 
to play these parts on a regular Bb trumpet, and even on the 
C or D trumpet they can be played only after gruelling 
practice. Occasionally the small Eb trumpet is brought into 
service in an attempt to master these high clarion parts. 
Many trumpeters bless Mozart and others for taking some , 
of these parts away from the trumpeter and giving them 
to the clarinetist. Mozart had a special "peeve" against the 
clarion, and even against the trumpet, for his sensitive ear 
could not endure the piercing, shrill voice of these instru- 
ments, especially in the upper register. Mendelssohn and 
other composers seemed to agree with Mozart that pity 

should be taken upon the trumpet players to say nothing 
about the audience and have arranged much of this clarion 
music so these extremely high trumpet parts have been 
eliminated. Conductors go still further today by delegating 
such high parts to oboes or clarinets in orchestras where the 
trumpeters are not equal to the task. 

Although some authorities assume that the clarions used 
to play these high and florid passages were all short instru- 
ments, others point out that some of this music could not 
be played on a short instrument because some of the notes 
written are not present on such an instrument. For instance, 
the German composer Steffani in 1695 composed a march in 
"I Trionfi del Fato" for four trumpets in C. Many of the 
notes in these parts are diatonic, some are chromatic, and 
they were written over one hundred years before the trumpet 
was made chromatic through the invention of either keys 
or valves. The only notes on trumpets of the time were open 
notes, such as are to be found on the bugle. As will be shown 
more in detail in later chapters, the successively higher notes 
on a bugle are increasingly closer together. For instance, 
the second open note is an octave higher than the first, while 
the third is only a fifth higher than the second. The fourth 
is only a fourth higher than the third, the fifth only a third 
above the fourth, and so on, until between the sixth and 
seventh there is only a step and a half. There is a whole step 
between successive notes until we reach the eleventh and 
twelfth, where only a half-step exists. This is also true be- 
tween the thirteenth to the sixteenth partials. 

It is obvious that if diatonic notes are to be played on a 
simple trumpet without valves, they must be played in the 
top part of the range, where the notes are close together. 

In order to play the notes written, it has been pointed out, 
it would be necessary for the player to use a trumpet which 
was eight feet long and which sounded as its fundamental 
the C two octaves below Middle C. It would then have the 
natural scale shown below, starting at No. I : 


No. 2 

On such an instrument it is clearly seen that the following 
music by Steffani could be played, because the notes neces- 
sary are to be found on such an instrument: 


If the instrument were only four feet long, having as it would 
a fundamental only an octave below Middle C, its natural 
scale would be as shown above, beginning at No. 2. It is at 
once obvious that there are gaps in such a scale which would 
make it impossible to play StefFani's music, D and F# being 
absent. Instruments of higher pitch, and consequently of 
shorter length, would have still greater gaps in the scale and 
would be even less adequate for playing this music. 

This is conclusive proof that some of the trumpets used 
for some of the high and florid solos of the late seventeenth 
and early eighteenth centuries were eight-foot trumpets. 
These instruments must have been very small in bore and 
must have used a small mouthpiece. Even so, it surely taxed 
to the utmost the powers of the players in those days to play 


such music. A survey of much of the music of these times 
shows that the first clarion player was regularly required to 
play up as high as the sixteenth partial. This is substantially 
the job of the French horn player, who is required to play 
up to the sixteenth partial on his horn. The notes on the 
horn, however, are not nearly so high in pitch, since the 
horn's fundamental is considerably lower than that of the 
eight-foot clarion, the horn being nearly thirteen feet long. 
Praetorius, writing about 1619, says the clarion in D was 
called upon to play from the seventeenth to the twenty-first 
partial. This range is almost unbelievable. The second clarion 
player was regularly required to play up to the tenth and 
eleventh partials. The third part for regular trumpet was 
written in the middle range, from the third or fourth to the 
sixth or eighth, the seventh being flat and seldom used. 

Praetorius says all trumpets of his time were built in D 
and that they were small in bore, short in length, and used 
a small mouthpiece. In this he is probably correct, except 
for the undoubted fact that eight-foot trumpets were often 
used. He was no doubt right in describing the bore, mouth- 
piece and pitch of the instrument as D, for this was generally 
the key in which they were built. They were also found 
occasionally in C and sometimes though rarely in G. Monte- 
verde in "Orfeo" scored for five differently pitched trumpets, 
but composers since his day did not follow this lead. When 
the trumpet became a regular member of the orchestra it 
was usually built in F, with crooks for changing the pitch 
to E, Eb, D, C, Bb and A. Sometimes trumpets were called 
for also in G, Gb> B, Ab and Db. 

While the trumpet was attaining great popularity in the 
orchestra, it was also gaining great favor with kings and 

princes. Today the rank of leaders of the nation is shown by 
the number of cannons fired in their honor. In former times 
the rank was shown by the number of trumpeters who accom- 
panied the royal personage. The royal bodyguard of Edward 
III was composed of seven trumpeters, while seventeen 
trumpeters saw to it that Charles II was given proper respect 
on all occasions. When the Prince of Orange journeyed from 
Holland in 1677, five trumpeters attended him. When the 
Earl of Mulgrave visited Tangiers in 1680, two trumpeters 
were assigned to attend him and signify his rank. Two 
trumpeters and one drummer were assigned to Prince Rupert 
in 1664. When these dignitaries alighted from their carriage 
to enter a building, warning of the approach of their excellen- 
cies was given by a fanfare of trumpets. When they boarded 
ship, the gangplank was cleared of common people by a 
blast from the trumpets. Whenever they walked in public 
thoroughfares, day or night, a flourish from the trumpets 
commanded respect for these distinguished personages. 

It was often common for drummers to accompany trum- 
peters in such duties. A roll on the drums and a fanfare, on 
the trumpets were calculated to arouse the proper respect 
for royalty. This association of the drum and trumpet was 
also found in the orchestra. Very often, when trumpets were 
called for in some of the earlier scores, drums were also used. 
In the early days of the orchestra, the trumpet and drum 
were associated in the composer's mind with things military. 
Usually when a military idea occurred in the composition, 
the trumpet and drum were called upon to play together. 
This was before the trumpet and drum were fully understood 
and before their wide range of colorings and effects was 

After Bach and Handel, trumpet playing declined. Haydn, 
the great successor of these two masters, did not do well with 
trumpets. When Haydn entered the service of Prince 
Esterhazy, music-loving prince of Austria, his orchestra at 
first did not include trumpets at all. As late as 1766, the 
regular personnel of this orchestra, one of the foremost in 
Europe, consisted of six violins and violas, one cello, one 
string bass, one flute, two oboes, two bassoons and four horns 
but no trumpets or cornets. Several years later the re- 
sources of the orchestra were enlarged so that trumpets and 
tympani could be added when needed. Even when Haydn 
did use trumpets, he scored for them so they played an 
octave or a sixth above the horns. To this thin arrangement 
he added drums for accompaniment. He probably felt the 
need of filling in with something, and the drums seemed the 
most appropriate. 

Mozart, who was at first Haydn's pupil but whose genius 
lifted him to a place above his master, seemed to share 
Haydn's dislike for trumpets. This antipathy for trumpets 
was due to an extremely sensitive nature. Until Mozart was 
ten years old, the sound of the trumpet was excruciatingly 
painful to him, and he could not endure it. As an adult he 
found little pleasure in trumpets, and he used them spar- 
ingly. In 1788 he wrote his three greatest symphonies, but 
in only two of them did he use the trumpet. He could not 
endure the high clarion parts written by Bach and HandeL 
He even rearranged some of this music, giving the high 
clarion parts to the clarinets. 

Beethoven generally wrote for two trumpets and often 
used them as solo instruments. This can hardly be inter- 
preted to mean that Beethoven was particularly fond of the 

trumpet, for it was a known custom of his to score as much 
as possible for all players in the orchestra and to pass around 
the solo parts in order to keep them all interested. In general 
he followed the custom of Mozart and Haydn in handling the 
trumpets, writing for them parts which were an octave, a 
sixth or sometimes a third above the horns, all to the ac- 
companiment of the pounding of the tympani. 

Although it probably was just as well that the trend was 
away from the high clarion writing of Bach and Handel, 
the composers who followed failed to invent any writing 
for the trumpet which was as interesting. Bach and Handel 
and their predecessors made the trumpet one of the most 
interesting instruments in the orchestra. They no doubt 
went to extreme lengths and exhausted the possibilities along 
this line, but they have to be given credit for resourcefulness 
and inventiveness. When composers after Bach and Handel 
abandoned this style of writing, they failed to bring forth 
anything to take its place. They used the trumpets much as 
bugles are used today in drum corps. The trumpet parts 
were thin chords whose poverty of design was covered up 
in the noise of the tympani. They apparently did not think 
well of the long trumpets on which it was possible to play 
chromatically in the upper registers. This kind of playing 
was a man killer for the trumpeters, but it did have possi- 
bilities which some feel were not fully exploited. These old 
masters also knew about adding crooks to the simple trum- 
pet, in order to obtain, by jumping from one trumpet to the 
other, something approximating chromatic playing. Wagner's 
success with this type of instrument shows well enough that 
Mozart, Haydn and Beethoven overlooked possibilities in 
the trumpet of their time. Instead of taking advantage of 

{ I7 o} 

the long trumpet with its diatonic and chromatic upper 
registers, and instead of using the trumpet with crooks as 
did Wagner, they contented themselves with writing thin 
tonic and dominant chords for these instruments. 

Possibly it is expecting too much, even from such geniuses 
as Haydn, Mozart and Beethoven, to look for trumpet 
writing beyond the thin chords based on the tonic and 
dominant. After all, although Wagner did great things on 
the simple trumpet without valves, he had set before him 
the example of piston-trumpet performance. He chose the 
simple trumpet because he preferred the tone to that of 
the valve trumpet, but the example of the valve trumpet 
must have suggested the superior trumpet writing for the 
simple trumpet. To appreciate what Haydn, Mozart and 
Beethoven were up against, we need only examine what sort 
of music is written for the regular military bugle today. 
Bugle calls are limited to five or six notes. Other notes are 
possible, but these five or six are the best in quality and the 
easiest to blow. Below are shown these notes (j^ 2 ), with 
other notes which are possible but seldom used ( J). The 
numbers indicate their relative position in the harmonic 
scale, or scale of open notes. 


The obstacles in making music with these notes are ob- 
vious. They have wide gaps between them, and their range 
limits the music to a monotonous span. In the upper part 
of the scale the notes are closer together and have greater 

musical possibilities, but these notes are hard to play and 
can be blown only by a few powerful individuals. Even with 
the accurately built instruments today, many players cannot 
hit the ninth and tenth partials; on the crude bugles two 
hundred years or more ago it is doubtful if many players 
could go beyond the sixth. It is little wonder that early 
composers did not think seriously about the musical possi- 
bilities of such instruments. 

It was to overcome such limitations and bridge the gaps 
in the natural scale, that the Kent bugle was invented in 
1810. This invention is often credited to James Halliday, 
an Irish bandmaster, and it became known as the Kent bugle 
in honor of the Duke of Kent, who became much interested 
in it. There is evidence, however, that Kolbel of St Peters- 
burg built a keyed bugle in 1770 and that Weidinger of 
Vienna also made such an instrument in 1801. These instru- 
ments were said to have five holes in the side which were 
opened and closed by five keys, similar in appearance and 
principle to the keys now found on saxophones. The in- 
ventors concluded, correctly enough, that they could make 
this instrument chromatic by using keys, just as the wood- 
winds had been made chromatic by the use of this device. 
Playing the keyed bugle was somewhat like playing the 
saxophone, a later invention, only the mouthpiece was cup- 
shaped instead of a mouthpiece using a reed. 

Immediately this invention was hailed as one likely to 
change the whole course of music. All brass instruments 
played with a cup-shaped mouthpiece could now play music 
just as a flute, clarinet or oboe. Instead of being neglected 
except for a few scattered chords, the trumpet and cornet 
could now take their place alongside the instruments which 

had been accepted for years in the orchestra and band. 
Abandonment of the high clarion writing of Bach and 
Handel was no longer regretted. Trumpet players would no 
longer have to choose between blowing their lungs out on the 
high notes or taking a back seat and playing only thin 
chords. They could now play their parts as did players of 
other instruments. A new voice, a new coloring had been 
added to the orchestra. Great prophecies were made about 
the benefits that would come to all the brass instruments as 
a result of this invention. 

The principle of using keys to bridge the gaps in the scale 
of cup-mouthpiece instruments was found to be better 
adapted to conical-bore instruments than to cylindrical-bore. 
For this reason, the keyed bugle became the most popular, 
with the keyed cornet less popular and the keyed trumpet 
not popular at all. This was a victory for the conical-bore 
instrument in its struggle with the cylindrical-bore trumpet. 
The demand for the keyed bugle increased suddenly, while 
the keyed trumpet dropped into the background. 

Unfortunately for the inventors, but fortunately for music, 
Blumel, an oboe player from Silesia, invented the piston 
valve in 1815, and the great keyed-bugle invention was 
quickly forgotten. Blumel may have obtained his original 
idea from an ingenious Irishman named Clagget, who in 
1788 rigged up two trumpets one in D and another in Eb 
jointed together so they could be played with one mouth- 
piece. By means of a change valve either trumpet could be 
connected with the mouthpiece. The advantage of this device 
is somewhat similar to that now on double French horns. 
By selecting the D or Eb trumpet, it was possible to obtain 
the open notes on either and thereby play a scale with fewer 

gaps in it than are found on a single trumpet. It is said 
Clagget also experimented with a valve similar to that de- 
veloped twenty-two years later by Blumel, but if he did, 
he did not develop it to a practical stage. 

BlumeFs device consisted of a sort of pump which could 
add a length of tubing to the main length, sufficient to lower 


i c 

] * v 



rn I 




t ii 




4 i 


3 L 


Showing the comparative lengths of the three valve slides of the 
cornet and trumpet. The first valve controls a length of tubing equal 
to a whole tone; the second, a semitone; and the third, a tone 
and a half, or three semitones. 

any open note a half-step. To the five or more open notes, 
such as were shown to be the only notes available on the 
simple bugle or trumpet, were added an equal number of 
notes through this piston device. This gave twice as many 
notes as before, and although the ability to flat all the open 
notes a semitone did not give everything desired on the 
trumpet or cornet, the principle was to be expanded so that 
all the gaps would eventually be bridged and the trumpet 
and cornet would be made truly chromatic. Blumel sold his 
invention to the German instrument maker Stolzel of 
Breslau. Stolzel carried the invention still farther by adding 
a second valve capable of adding a length of tubing to the 

main length sufficient to lower the tone of any open note a 
full tone. The two valves could be combined to add a step 
and a half. In 1830 Miiller of Mayence saw the need of a 
third valve and added one controlling a length of tubing 
capable of adding three semitones. This is the modern 
arrangement of valves and makes the scale completely 
chromatic above the second partial or open note. 

To understand how this valve worked we need only take 
the bugle scale, which is the scale for all simple horns. If we 
start with the G above the staff, we can bridge the gap to 
E below by adding first the half-step tubing (bringing us 

2 1 

Open notes are shown as half-notes and are playable without 
valves. Notes played by using the valves are shown as quarter- 
notes. Numbers under notes indicate which valve or combination 
of valves is used, the valves being numbered starting with valve 
nearest mouthpiece. The numbers in parentheses indicate notes 
not playable on two-valve instrument, but made playable after the 
third valve was added in 1830. 

down to F#) and then the whole-step tubing (bringing us 
down to F). The next note is open, so we play chromatically 
from G to E by pushing down the half-step valve, then the 
whole-step valve, and then playing E open. To go on down 
to C chromatically is done in the same manner: push down 
the half-step valve and sound D$, then push down the whole- 
step valve and play D; finally push down both valves, 
thereby adding to the open E a step and a half, and C$ will 
sound. Since C is open, the next half-step is reached without 
any valve. 

With the two valves it is not possible to bridge the gap 
between this open C and the open G below, for it is obvious 
that the two valves and their corresponding tubing can take 
us down below C only a step and a half, leaving us stranded 
on At]. It is not possible to play G#. Also, there is a gap be- 
tween the open G and the open C below, for it is possible 
to play down from G chromatically to El?, leaving a gap of 
D$, Dtj and Cfr But the third valve was added, controlling 
one and a half steps. By pushing down all three valves and 
using the tubing of all three, six semitones are made avail- 
able. These are sufficient to bring us down chromatically 
from open G to the open C below. This makes the entire 
effective range of the trumpet and cornet chromatic. Of 
course, the fundamental of the instrument is an octave below 
this lowest open C, but this is impossible for all but a few 
artists to play. By using the six semitones, it is possible to 
descend below this open C to F$ below the treble clef, but 
these notes are not very musical at best and are obtainable 
only by a few players with unusual lips. This still leaves a 
gap between F$ and the C below, but this does not present 
any problem to the trumpet and cornet player, for he does 
not care to go down that far anyway. This problem does 
present itself to the tuba player, and a fourth valve has been 
added, as will be related in a later chapter. 

It would be supposed that composers would seize upon the 
new valve trumpet immediately and would score for it in- 
tensively, if being a newcomer to the orchestra and conse- 
quently offering new resources never used before. At least it 
would be supposed that the new valve trumpet would out- 
mode the old simple trumpet. Such was not the case, at 
least not immediately. It was twenty years before the valve 

trumpet was called for to any extent. Beethoven and his 
contemporaries seem to have ignored it. With a few rare 
exceptions, there are no trumpet notes written in the scores 
of this time which could not be played on the simple trumpet. 
It is possible that even these were written for the keyed 

It is surprising how the simple trumpet with crooks hung 
on after the valve was invented. Scores called for the trumpet 
with crooks until nearly 1900, but the F trumpet and the 
Bb-to-A trumpet gradually became standard. The chief 
crooks used were C, D, Eb E and F, those common in the 
music of Mendelssohn and Schumann. Other crooks were 
called for, in practically every key from low At to high C. 
The valve trumpet is said to have been specified in a score 
for the first time in 1836 by Meyerbeer in "Les Huguenots/* 
Another contender for the honor is Chelard, whose "Mac- 
beth" was performed in Paris in 1827 and whose score is 
said to call for valve trumpet. With Schumann, Wagner and 
Berlioz, the valve trumpet became rather generally accepted, 
so that by the latter part of the nineteenth century it was no 
longer necessary for the composer to specify " valve trumpet" 
or "piston trumpet" to distinguish from the simple trumpet, 
the word trumpet meaning by then the trumpet with pistons, 
or valves. 

In England the valve trumpet had a difficult time in gain- 
ing any acceptance, owing to the popularity of the slide 
trumpet. This instrument was hardly known outside of 
England, but it had a tremendous vogue for about a hundred 
years, beginning with the close of the eighteenth century. 
Some authorities have confused this instrument with the 
treble trombone, but it is distinctly different. It was usually 

built in F, the same pitch as the classical F trumpet. Coming 
onto the scene before either keys or valves had been added 
to the trumpet, it was of course a simple trumpet, but it 
had a slide which was capable of lowering any natural, or 
open, note a half- or a whole tone. This slide was something 
like a slide on a trombone, except it was shorter and worked 
in an opposite way. Instead of being pushed forward to 
lower the pitch of a note, it was pulled backward toward 
the player. Its position was approximately where the first 
valve slide on the modern valve trumpet is now located 
and was rigged up with a spring so that when the slide was 
released after being pulled out to flatten a note, the slide 
jumped back into closed position. 

Although there were some famous performers on the slide 
trumpet, it was difficult to play. It could add only a half- 
tone or whole tone to the open notes, which was not sufficient 
to make the trumpet chromatic, any more than a trumpet 
with a half-step valve and a whole-step valve could be 
chromatic. Serious gaps were left in the scale which damp- 
ened the ardor of its many admirers. The intonation was 
bad, too, especially when the trumpet was played with 
crooks. When played in F, the intonation was fairly good, 
but when crooks were added, the slide was not long enough 
to add a whole step to the open notes, because of the longer 
interval required for a whole step in the lower reaches of 
the scale. When the trumpet happened to be in higher 
pitches, such as Bb, C or D, the positions of the slide were 
so radically altered that only the most skillful players could 
play the trumpet in tune. Although the slide trumpet de- 
layed for over fifty years the introduction into England of 
the more efficient valve trumpet, eventually, along toward 



the close of the nineteenth century, the valve trumpet began 
to displace the slide trumpet, and after 1900 it was mostly 
a historical relic. 

Wagner is given credit for being the first composer to 
score well for the trumpet. He usually used three and some- 
times he used four trumpets. He became acquainted with 
their resources, appreciated their colorings and conceived 
effective ways of scoring for them. It was a long time, how- 
ever, before Wagner abandoned the old custom of scoring 
for trumpets with crooks to throw them into various keys. 
There was a belief in those days that trumpets in different 
pitches possessed different qualities of tone. This is difficult 
for us to understand, because today the trumpet player uses 
the Bb or C trumpet and transposes all music written for 
variously pitched trumpets. Nobody seems to think that 
any desirable tonal coloring is lost by this method. Possibly 
the trumpets in Wagner's day were not nearly so accurate 
in intonation, and it may be there was some justification 
for his practice. One thing is sure, his method of writing 
worked a severe hardship on the trumpet player. A classical 
example is the ninety-six bars in the opening of the intro- 
duction to "Lohengrin," where ten crook changes are 
marked. The rate at which these ten changes must be made 
is indicated when it is considered that only two and a half 
minutes are consumed in playing these ninety-six bars. 
This averages a change of crook every fifteen seconds. 
Trumpet players began to revolt against such practices. 
They found they could easily transpose the music, and they 
preferred to transpose rather than to change crooks. This is 
the universal practice today. 

Following the invention of the keyed bugle in 1810, keys 


were added to both cornets and trumpets. It was not long, 
however, until piston valves were installed in both cornets 
and trumpets. Although trumpets were much the more 
popular, cornets were scored for to some extent, especially 
by French composers, notably Berlioz and Bizet. This in- 
strument was known as the "cornopean." Usually they 
were called for in addition to trumpets, and not as trumpet 
substitutes. If they did not exclude the trumpets, they did 
all but obliterate the old keyed bugle and later the piston 
bugle. In the middle of the nineteenth century this instru- 
ment had quite a vogue in the form of the well-known sax- 
horn. The saxhorn was the invention of Antoine Sax in 1842. 
There were seven instruments in the family, from a high 
soprano in Eb to a contrabass in B[?. In bore they were all 
conical, and their tone was especially mellow and of singing 
quality. These instruments are still used in some of the 
European brass bands, but they are rare in America. After 
inventing the saxhorns, Sax invented another family of in- 
struments, called the saxtrombas. Instead of having a conical 
bore, they had a bore more like that of the cornet; that is, 
they were only conical in part of their length, the rest of the 
bore being cylindrical Although there are many who regret 
that the saxhorns did not survive, rather than the saxtrom- 
bas, the brass instruments in use today are more like f the 
saxtrombas than the saxhorns. Those who prefer the sax- 
horns say the saxtrombas are so nearly like the trumpets 
and trombones as to lack contrast necessary for best musical 

Although in the larger horns, such as baritone and bass, 
only the saxtromba bore has survived, in the soprano instru- 
ments we have retained one of each type. We retain the old 

1 180.1 

trumpet with its cylindrical bore, shallow cup mouthpiece 
and bright tone. This is the brass soprano of the orchestra. 
We also retain the cornet, with its partly cylindrical, partly 
conical bore, its deeper cup mouthpiece and its mellow 
quality of tone. This is the brass soprano of the concert band. 
The more nearly conical in bore bugle and saxhorn have 
survived in the fluegelhorn. Its bore is conical in shape, its 
mouthpiece is larger and deeper than that of the cornet, and 
its tone is broader and more mellow than that of the cornet. 
Some have called the fluegelhorn a mezzo-soprano, and that is 
really what it is. It is found in the band and serves to blend 
the tones of the cornets and trumpets with the altos and 
French horns. 

The classical trumpet was built in F, a fourth below the 
Bb in which trumpets are usually built today. Being lower 
in pitch, this instrument had a more beautiful lower register. 
On the other hand, its top register was not as good as that 
on the Bb trumpet. As the tendency was to write higher and 
higher for the trumpet, the Bb instrument came to be the 
standard pitch for both trumpet and cornet. This tendency 
to write higher and higher for the trumpet has led to the 
use of the C trumpet, and even trumpets in D and Eb. 
Occasionally brass bands call for the Eb cornet also. 

Although built in Bb, most trumpets and cornets have a 
slide which can be pulled out, bringing the pitch of the in- 
strument down to A. The majority of compositions for band 
are written in flat signatures, and such signatures are easier 
to play by using an instrument built in a flat key. For in- 
stance, a composition written in three flats has only one 
flat for the Bb trumpet. Orchestra music, on the other hand, 
is usually written in sharp signatures. This makes it helpful 

to use an instrument built in a sharp key. For example, a 
composition written in three sharps has a part in C (no 
sharps) for the A trumpet. This explains why trumpets and 
cornets are built in Bb with slide to A. 

Although players often refer to "blowing" the cornet or 
trumpet, such instruments are not "blown" at all. If a 
person picks up a trumpet and blows into it, no sound comes 
out. The instrument is made to sound by setting up a vibra- 
tion of the air inside the tubing. This vibration is made by 
pressing the lips against the cup-shaped mouthpiece and 
blowing between them. Teachers often describe the lip 
action as that required by spitting a hair from between the 
lips. The weak buzzing sound produced by the lips when 
air is forced between them is picked up by the instrument 
and amplified, as will be shown more fully in closing chapters. 

Although the orchestra has become almost classical in 
most respects and has changed little in recent years, the 
American concert band has been undergoing great changes. 
The old brass band is practically a thing of the past, and 
in its place is evolving the concert band. This has had an 
important bearing on cornets and trumpets. In the old days 
the soprano section of the brass band was composed almost 
entirely of conical-bore saxhorns and later of cornets. The 
altos and tenors and baritones and basses were also conical- 
bore instruments of the saxhorn family. Gradually wood- 
winds found their way into the band. More and more were 
admitted, and the brass was pushed back to give room. 
Eventually there were more clarinets in the band than cor- 
nets or saxhorns. At first the clarinets were grouped to the 
left of the director, with the cornets on the right. Lately, 
however, first clarinets have been stationed on the left of 

the director, just as are the first violins in the orchestra. 
On the right of the director are second clarinets and other 
woodwinds, such as bassoons, oboes, alto and bass clarinets. 
Back of these, with the trombones, will be found a group of 
cornets and trumpets. 

Following the era of florid cornet soloing by such virtuosi 
as Levy, Liberati, Hoch, Smith and Bellstedt, the cornet 
band became popular. It was seldom that a trumpet could 
find a foothold in the band. Gradually the trumpet came to 
the front and bands were filled with trumpets, almost to the 
exclusion of cornets. Along with the growing popularity of 
the woodwinds has come a realization that cornets are 
better fitted to be the brass sopranos of the concert band 
than trumpets. The American Bandmaster Association has 
specifically recommended that the cornet be used in the band 
instead of the trumpet. There is no objection to having a 
couple of trumpets in the band to give contrasting coloring 
for effects, just as occasionally two or three cornets will be 
used in the orchestra for effects; but the cornet has very 
definitely become the principal soprano of the brass choir 
of the band. Since there are several times more soprano 
voices used in the brass choir of the band than there are 
used in brass choir of the orchestra, the cornet can be said 
to have achieved an edge today on the trumpet in the age- 
old feud which has existed between them for several cen- 

Although the differences among trumpet, cornet and flue- 
gelhorn are generally accepted to be as given in the foregoing 
pages, actually today these differences have become greatly 
minimized in manufacture. The generally accepted defini- 
tions that the trumpet is two-thirds cylindrical and one-third 

conical, while the cornet is one-third cylindrical and two- 
thirds conical, are really not true. Actually the only part of 
the trumpet tubing which is straight is the valve and tuning 
slides, but these likewise are the only straight tubing on the 
cornet. The mouthpipe of the trumpet used to be mostly 
straight, but today even it is generally tapered. The actual 
sliding part of the tuning slide is cylindrical, but on most 
trumpets the side joining the mouthpipe is smaller than the 
side leading to the valves, thereby giving it a general taper. 
It is distinctly true, however, that the degree of taper is 
different among the three instruments. There is less taper in 
the trumpet, more in the cornet, and most in the fluegelhorn. 
The mouthpiece receiver on the trumpet is larger than the 
receiver on the cornet, while the bell is usually smaller, 
thereby allowing more taper between mouthpiece and bell 
in the cornet than in a trumpet, both of the same bore 
through the valves. Therefore, although the two-thirds-to- 
one-third relationships do not hold, actually there is an 
effective difference among them which still defines their 
tonal qualities. 


The French Horns 


FIVE GENERATIONS OF PEOPLE existing at one time call 
for a photograph and a write-up in the local newspaper. 
Sometimes a span of ninety or a hundred years is covered 
by such a group of prolific and Iong4ived persons. The 
modern French horn wins all the honors for such distinctions 
among musical instruments, for the horn's remote ancestor 
the Hebrew shofar stands beside it today, hale and 
hearty after sixty centuries of recorded history. 

The Hebrew shofar is the oldest type of horn of which we 
have any authentic record. It was used in religious ceremonies 
by the Hebrews six thousand years ago and is still used 
today in Jewish ceremonies connected with the New Year 
and the Feast of Atonement. No other instrument can point 
to such a long record of continuous use. Hundreds of genera- 
tions of rams have furnished horns from which the shofar 
has been made, but during all these ages this ancient instru- 
ment has not changed. It is the same today as it was six 
thousand years ago when the fathers of Abraham introduced 
it into their religious ritual and ceremony, even to the sharp 
rim of the cup serving as a mouthpiece. 

The shofar is made by cleaning out the inside of the horn 
of a ram and making a sort of funnel or cup-shaped opening 
at the small end to serve as a mouthpiece. This cup often 
has a sharp edge, as has been found out by a number of 
today's French horn and trumpet artists who have tried to 
blow this instrument. Their lips are accustomed to the wide, 
smooth rim of a metal mouthpiece, and when they participate 
in the Jewish services, as they sometimes are called upon to 
do, they find the sharp edge of the ancient shofar most 

Of course instruments were made from the horns of ani- 
mals a long, long time before the recorded history of the 
Jewish race. All primitive peoples made and still make in- 
struments from the horn of rams, cows, buffaloes. Tusks of 
the elephant are also a favorite material This crude kind of 
musical instrument no doubt is the common ancestor of all 
cup-mouthpiece instruments, including trumpet and trom- 
bone with cylindrical bore, but the French horn is the true 
son of the ancient instrument made from an animal horn, 
because it is conical in bore like the animal horn. The 
natural horn without valves carried a taper from the mouth- 
piece to the bell, and the valve horns of today are tapered 
throughout except through the valves. 

It is conjectured that the Queen of Sheba was greatly en- 
tranced by the shofar when she visited King Solomon and 
carried many of these instruments back home with her; in 
any event, the ancient Ethiopians used a similar instrument 
in their religious ceremonies which was known as the kenet. 
The Greeks also used the kegas to announce sacrificial 
offerings and to call the people together for various religious 
and secular gatherings. Alexander the Great is said to have 


used a great horn for assembling his troops, and tradition 
has it that such a mighty blast could be blown upon it that 
it could be heard for two miles. A similar horn is said to 
have been used by Roland, famous warrior nephew of 
Charlemagne. When leading a division of Charlemagne's 
great army, Roland upon several occasions used this horn 
to signal his uncle, miles away, for assistance when the tide 
of battle turned against him. 

Chaucer furnished an early record of the horns in England. 
There are many passages in his writings which refer to the 
"bemes," the term for horns. In one passage he implies that 
they were made of many materials but were not any too 

"Of bras thay broughtm bemes and of box \wood} y 
Of horn 9 of boon [bone], in which thay blewe and pouped. 
And therewithal thay shryked and thay houped." 

Such horns were used mostly in war, in the chase, and for 
different kinds of signaling. Henry VIII seemed to consider 
them genuine musical instruments, for he left twenty-two 
horns of various kinds in his collection of musical instru- 

It was in France, though, that the horn met its great 
opportunity and where it started its rise into the ranks of 
orchestral instruments. It is because of the prominence given 
the horn in France that it has come to be known as the 
French horn. It started its apprenticeship as a hunting horn, 
but served it so well it was finally accorded a place in the 
orchestra. It found favor with the French king Louis XI, 
who reigned in the latter part of the fifteenth century. At 
that time the horn was made of metal and was bent in a 

circular shape, inscribing about three fourths of a circle. 
The mouthpiece was deep and shaped like a funnel, as are 
the horn mouthpieces of today. Being short in length, the 
horn had fewer notes than it possessed later. Its repertoire 
consisted of a few short blasts which served as simple 
signals. Louis XIII, who ruled during the first half of the 
seventeenth century, used a longer hunting horn which was 
curved in circular form and described a complete circle with 
some overlapping. Having more possible notes, the horn was 
given a larger repertoire. Short, simple notes in a series were 
sounded instead of single blasts, to signal certain information 
to the hunters. One of the most elaborate was an abbreviated 
melody of several notes which indicated that the fox had 
been apprehended. 

These calls seemed to please the French kings and their 
nobles, and new calls were added and old ones made more 
elaborate. This quest for more musical calls resulted in 
making the horn longer and longer, to afford more and more 
notes. By the middle of the eighteenth century, Louis XV 
was using a horn consisting of three complete circles of 
tubing, the instrument encircling the body and resting on 
the shoulder. Many complicated calls were used, and it 
became necessary for the hunt master of Louis XV to sys- 
tematize the growing repertoire of calls. The shorter, more 
simple calls were reserved for such ordinary tasks as cheering 
the hounds, signifying the progress of the hunt and asking 
for aid in case of emergencies. A second group of more elabo- 
rate calls was used to signify the kill. They were so systema- 
tized that one call indicated a deer, another a fox, and so on. 
They could even indicate the color of the animal, the number 
of antlers and other descriptive information. These calls 



were a kind of telegraphic code set to music. The third group 
of calls was the most elaborate of all. These calls were in 
reality musical airs. They were played after the hunt was 
over as a means of celebrating and rejoicing. Many of them 
were of a high musical order and later were incorporated in 
orchestral scores much as they were played originally in 
the hunt. It was this group of horn calls which bridged the 
way for the entrance of the lowly horn of the chase into the 

Even before Louis XV, the hunting horn had impressed 
certain composers well enough that they wrote orchestral 
parts for them. Lully, an Italian by birth, is given credit for 
being the first to do so. He wrote parts for the "trompes de 
chasse" in his "Princesse d'Elide," performed in Paris in 
1664. Apparently this was done as an experiment, and that 
Lully was not any too well pleased with the results is indi- 
cated by the fact that he didn't continue using the hunting 
horns. These instruments did not become regular members 
of the orchestra until nearly a hundred years later. When 
Rameau used two horns in his "Zoroastre" in 1749, the 
hunting horns were coming into their own as orchestral 
instruments. Rameau used these two horns in a kind of duet, 
which consisted partly of hunting calls and partly of sus- 
tained harmony passages. Grossec scored two horns to play 
an obbligato for some arias written in honor of Sophie 
ArnoukTs debut in Paris in 1757. 

Although the horn was first introduced into opera in 
France, it was in Germany that it first attained prominence 
as an orchestral instrument. Lully was bold enough to try 
the hunting horn in 1664, but it was not seriously considered 
in France until nearly a hundred years later. In Germany, 

however, it became an instrument to reckon with early in 
the century. It was known in this country as the waldhorn, 
or "forest horn," and was first used in the orchestra by 
Keiser in his opera "Octavia," performed in Hamburg in 
1705. This composition called for a pair of horns in C. Six 
years later two horns became regular instrumentation in the 
orchestra of the Royal Theater of Dresden. Not to be out- 
done, the Imperial Opera orchestra in Vienna the following 
year added two horns. From this time on the horn was a 
popular instrument in Germany, especially in the northern 
part, and a historian writing in 1713 about musical instru- 
ments in Germany said the horn was in considerable demand, 
not only in theater orchestras, but also in small ensembles 
and in churches. In 1721 Bach called for two horns in his 
first "Brandenburg Concerto," and the position of the horn 
was assured. 

To Bach's countryman and contemporary, Handel, goes 
the honor of introducing the horn into England, in 1715. 
This event is not as English as it might seem, for Handel, 
the English king and the horn were all three from Germany. 
Handel was chief musician to the Elector of Hanover, in 
Germany. He secured permission from the Elector to go to 
England for a short visit, but once in London, he tarried 
many months, which greatly irked the Elector. At the end 
of two years the Elector of Hanover was made king of 
England, much to the embarrassment of Handel, who was 
still in England. As a means of gaining back the favor of his 
former German patron, now English king, Handel composed 
his "Water Music." It was in this series of compositions that 
Handel introduced the German waldhorn. Apparently his 
efforts were pleasing to George I, for he welcomed Handel 

[ 190} 


back into his favor* Five years later Handel used two horns 
in D, in the London performance of "Radamisto." From 
that time on, the horn's place in the orchestras of England 
was accepted, for every great composer since Handel's day 
has used this instrument. 

The horn was also used in Italy by the early composers of 
opera. Scarlatti seems to have been one of the first Italians 
to appreciate the horn, for in 1715 he wrote parts for two 
horns in F, in his opera "Tigrane," performed for the first 
time in Naples. He wrote mostly in the upper register of the 
horns, in order to have at his command sufficient notes to 
compose simple melodies. Later as he learned more about 
the use of horns he wrote harmonies more in the modern 
style of composing. 

The horns of this early period, up to the time of Bach and 
Handel, were played with the bell up. The mouthpipe was 
horizontal with the mouth, then the tubing made a couple 
of turns, ending with the bell opening straight up. The 
Germans were responsible for reversing the position of the 
horn, turning the bell downward to its modern position. 
This position was arrived at in their efforts to subdue the 
harsh tone of the horn by partially stopping it with the hand. 
In order to do this the bell was turned downward so it was 
possible to place the hand in the bell. They also muted the 
horn with mutes made of wood and cardboard. 

The horn at this time was about seven and a half feet 
long, and various pitches were obtained by inserting crooks 
of various lengths in the mouthpipe. Some of these crooks 
were longer than the main length of the horn itself, the 
longest being twelve and a half feet. Even short crooks 
Inserted in the mouthpipe threw the intonation off, and it 

can be imagined how terrible in intonation were the horns 
with these long crooks. At first horns in D and F were mostly 
used. Later horns in C and G were common. Still later crooks 
were added to throw the horns into Eb, E and A. Then 
longer crooks were added to give low Bb and low Ab. Short 
horns were also constructed to give high B and even high C. 
In 1755 Johann Wernern, second horn of the Imperial Opera 
orchestra of Vienna, used what was called the "improved'* 
horn. It had a total of nine crooks, giving him horns in the 
following keys: low Bb, C, D, Eb, E, F, G, A and high Bb. 
Even this was not enough for some horn players, and it is 
said that they finally packed around with them a total of 
sixteen crooks, ranging from low Ab to high C. To divide 
this great number of crooks among several horns was bad 
enough, but to supply sixteen crooks with one horn was 
almost criminal. Such a horn could not possibly be made to 
play in tune. 

It is no wonder that horn players revolted against such 
practices. Hampel, a horn player in the Dresden orchestra, 
was the first to do anything about it. He schemed around 
and finally discovered that less serious difficulties would be 
experienced if the crooks were inserted in the body of the 
horn, rather than in the mouthpipe. In 1753 he brought 
out his celebrated "Invention horn," also called the "Ma- 
chine horn." This horn was devised so that movable slides 
could be inserted in the body of the horn, somewhat as 
tuning slides are inserted in the horns of today. Acting in 
self-defense, Hampel did a great service for his fellow horn 

Hampel was a man of genius and continued his experi- 
ments. He noted that oboe players were accustomed to 


inserting a wad of wool in the bell of their instruments to 
soften the harsh tone. Horns in those days were not any too 
sweet and smooth in tone, and he conceived the idea that 
this trick might do the horn some good, too. What he found 
out led to a new era of horn playing. He found that this de- 
vice not only softened the tone of the horn but raised the 
pitch a semitone. Later the wad of wool was abandoned and 
the hand was inserted into the bell of the horn. By raising the 
pitch of the horn a semitone, it was possible to sharpen by 
a half-step all the open notes in the natural scale, thereby 
adding a complete new series of notes and making the horn 
chromatic in the upper part of the scale where the open notes 
were close together. This discovery was made in 1760, and 
although used quite extensively by most horn players, com- 
posers did not write regularly for the "hand" horn until 
early in the nineteenth century. 

Other attempts were made to fill in the gaps of the open 
scale. Notable among these was the invention of Kolbel, 
a Bohemian horn player at the Chapel Royal of Russia. 
Kolbel in 1770 conceived the idea of placing a key on the bell. 
By opening this key, the effective length of the horn could 
be reduced sufficiently to raise the pitch of all open tones a 
semitone. The idea seemed to win favor, and by 1801 the 
number of keys had been increased to five keys, making the 
horn chromatic, except between the two lowest open tones. 
A few years later the sixth key was added, giving a chromatic 
scale from low C to the top of the scale. Only the bell series 
of notes was of good quality, however; the other series of 
notes came from the holes and were not as clear and resonant. 
The difference in quality was generally recognized and led 
to the practice of detaching the bells altogether. The horn 

without the bell produced a harsh tone, and although not 
desirable musically, the practice was adopted more or less 
generally for certain effects. Composers wrote notations for 
such passages specifying "bells off." An instance of this is 
found in MehuPs duet for horns in "Euphrosyne et Coradin." 

Horn players can thank the oboist for the invention of the 
stopped horn, as already noted. They can also thank the 
oboist for another and even greater improvement to their 
instrument the valves. In 1815 Blumel, an oboe player of 
Silesia, invented the piston valve. By this device, a length 
of tubing could be added to the length of the horn by pushing 
down a piston, or pump. The length of tubing so controlled 
was sufficient to add a half-step to the open notes, giving a 
whole new series of notes. The ease with which this could be 
done appealed to horn players and also to the players of other 
cup-mouthpiece instruments. The invention was sold to 
Stolzel, a German, who refined the invention and added a 
second valve which controlled a length of tubing sufficient 
to lower all open tones a whole step. In 1830 Miiller of 
Mayence added a third valve which operated a third loop of 
tubing sufficient to lower the open tones one and a half 
tones, making the instrument chromatic. Ten years later 
Stolzel's inventive genius produced the sub-ventil valve, 
so called because the valve tubing came out at the bottom 
of the valve. This was hailed as the last word in valve con- 
struction, but modern instruments have gone back to the 
valve whose valve slides come out at the side. 

Blumel was not satisfied with the piston valve, believing 
that the travel of the piston was too long and clumsy. He 
finally hit upon the rotary valve in 1827, This valve con- 
sisted of a little rotor or drum which revolved on an axle. 

Cross sections, showing the difference between the French-horn 
mouthpiece (left) and the mellophone mouthpiece (right). Note 
that the horn mouthpiece is deeper and more funnel-shaped. 1 his 
difference in shape has much to do with difference in tone quality. 

The Hebrew shofar, still used in religious 
services after sixty centuries. 


Slots were milled in the side of the rotor so that when it was 
turned the air column was switched into valve slides, thereby 
adding various lengths of tubing for lowering the open tones. 
It was another ingenious device for accomplishing the same 
effect as that of the piston valve. Although the piston valve 
has survived in France and England, the rotary valve is 
popular in Austria, Germany and also in Italy. In America, 
the rotary valve has found favor on only the French horn, 
most French horns using this type of valve. The action of 
the rotary valve is shorter, faster and more delicate than 
that of the piston valve, and fine French horn players almost 
universally prefer this type of valve to the piston type. 

Although the advantages of the valve horn are obvious, 
its adoption was long delayed because of custom and the 
general reluctance of musicians and composers to adopt any- 
thing which is different. It was twenty years after the in- 
vention of the valve before it was generally specified. Bee- 
thoven, Weber, Schubert and Rossini used the old hand horn 
freely, and it was not until about 1835 that the valve horn 
was really accepted. Even then it was the custom to use two 
hand horns with two valve horns. There was a feeling that 
valve notes were not as good as the open notes on the hand 
horn. After valve horns were generally used, composers often 
called for hand horns with crooks for certain passages where 
they wanted to display the French horn quality of tone to 
the best advantage. The valve horn did not win the battle 
over the old hand horn until toward the latter part of the 
nineteenth century, and it was not until the final quarter of 
the nineteenth century that the hand horn was entirely 
abandoned in "favor of the valve horn. 

Today it is customary to use four horns, but when the 

horn was gaining a foothold in the orchestra composers were 
undecided just how many were needed to produce the best 
effects. When Bach wrote the first of the "Brandenburg 
Concertos" in 1721, he called for only two horns. He seemed 
to be following the practice used in writing for oboes and 
bassoons, because a pair of these was usually used. Haydn 
in 1759 used two horns, along with two oboes, in his First 
Symphony, and we find he was still using two horns when he 
composed his D Symphony thirty-six years later. It is true 
that in 1766 his famous orchestra at Eisenstadt included four 
horns, but this orchestra was distinctly an opera orchestra 
and not a symphony orchestra. 

Mozart's three great symphonies written during the sum- 
mer of 1788 all call for two hand horns. Three years later 
when he wrote the opera "Idomeno," he employed four 
horns. In this he was merely following the custom of the time. 
Two horns were used in symphony, and four horns were often 
used in opera. Usually two of the horns were in one key and 
two were in another key, giving in this way more open notes 
than if all four were in the same key. The melody or harmony 
was passed from one pair of horns to the other in order to use 
the notes available on them. Generally Haydn and Mozart 
wrote from the second open note up to G or A above the 
staff, the twelfth or thirteenth note in the open scale. Oc- 
casionally they wrote up to high C, using the fourteenth, 
fifteenth and sixteenth open notes, but this depended upon 
the key of the horn, for these extremely high notes were not 
possible on the horns of high pitch. Although hand stopping 
had been discovered some time before, only a few stopped 
notes were called for. 

Beethoven started out writing for two hand horns, as in 


his C Major symphony composed in 1800, but five years 
later when he composed his great "Eroica" symphony he 
called for three horns. This is the first time three horns had 
ever been used in the symphony. The fourth horn was added 
to the symphony in 1824, when he called for a quartet of 
horns in his Ninth Symphony. That four horns were an 
innovation not generally accepted is shown by the fact that 
Schubert, Schumann, Mendelssohn and others who im- 
mediately followed Beethoven used sometimes two horns 
and sometimes four. These composers were obviously think- 
ing in terms of two instruments, for they generally gave to 
the two horns two notes, to form a chord with two bassoons, 
giving the lower notes of the chord to the latter two instru- 
ments. When they did write for four horns they usually gave 
the two upper notes of the chord to the first and second horns, 
and the two lower notes to the third and fourth horns. It was 
not until later that they fell in with the custom, almost 
universally adopted now, of giving the two top notes to the 
first and third horns and the two lower notes to the second 
and fourth horns. 

All composers have not adhered to this rule, however. 
Various French composers of the nineteenth century, among 
them Saint-Saens, very often had the third and fourth horns 
play the higher notes and gave the lower range to the first 
and second. Recently composers have considerably changed 
the writing for French horns, and this tendency is becoming 
more and more common. They are learning to use them in 
rotation, 1-2-3-4. I* 1 many instances, the doubling of the 
first horn is done by the second and the doubling of the 
fourth horn is done by the third, just opposite to the custom 
adopted by the classic composers. The second horn of today 

must have a good upper range, and the third horn must have 
an easy and good lower range. 

Although these composers wrote during the era of the 
valve horn, they almost always used the hand horn. Schu- 
mann was one of the first of the great composers to use the 
valve horn; for instance in his Third and Fourth symphonies. 
Although Schumann was not as good an orchestrator as 
Schubert and Mendelssohn, it is to his credit that he saw, 
better than they, the value of this new invention, 

Rossini did a great deal for horn playing by demonstrating 
its possibilities as a solo instrument. He experimented with 
highly florid passages and showed that the horn was entirely 
capable of this type of playing. He undoubtedly enjoyed 
advantages not possessed by his contemporaries, for his 
father was an expert horn player, and he was brought up on 
a diet of horn music. He knew the horn intimately and was 
a recognized authority on the instrument. He is reported to 
have rebuked his father for bad performance one time while 
his father was playing under his direction. The orchestra was 
rehearsing one of Rossini's operas in a small town in Italy 
when his father hit a "sour" note, Rossini looked over 
gravely at his father and said, "You may go right home!" 

Weber is celebrated for his beautiful handling of the horns, 
as in "Oberon" and "Euryanthe." One of the most famous 
passages for the horn is the noble horn quartet in "Der 
Freischiitz." Meyerbeer demonstrated new possibilities for 
the horn in the masterful way in which he delegated melody 
and thematic material to four horns. Berlioz believed a good 
way to treat the horns was to pitch them in four different 
keys, thus affording a great number of open notes. He recom- 
mends this form of writing in his famous Instrumentation and 


Orchestration. He also makes a plea for the natural horns 
without valves, but toward the close of his treatise on horns 
he has to admit that the valve horns have advantages over 
the old hand horn. Berlioz, along with all other Frenchmen, 
was backward in his attitude toward the old hand horn. 
After all other nations had discarded the hand horn, France 
was still using it, but finally had to give in and adopt the 
valve horn. 

Wagner followed the practice of his time by using both 
hand and valve horns, using the latter for first and second 
parts, and the former for third and fourth. Not until "Lohen- 
grin/' finished in 1848, did he abandon the natural horn for 
the valve horn. In the "Ride of the Valkyries," finished in 
1856, he wrote for eight horns in an effective manner. Using 
eight horns probably suggested the need for carrying the horn 
harmony down lower in the chord, for we find him advocat- 
ing a sort of bass French horn. He had a quartet of these 
made, called tubens. There was the tenor in Bb and the bass 
in F. They were of long, small tubing and were played with 
a funnel-shaped mouthpiece. Such instruments, however, 
do not achieve the true French horn quality and never be- 
came widely used. Horn players say they are very difficult 
to play in tune but that they respond easily. A very tight 
embouchure must be used because of this freedom of re- 
sponse. To make a bad situation even worse, Wagner wrote 
for two pairs of tuben in the opposite key. Parts for the tuben 
in F were written in Bb and parts for the Bb tuben were 
written in F; and both parts were written in the bass clef! 
One of the very first to use the tubens was the Chicago 
Symphony Orchestra, under the celebrated Theodore 
Thomas. Max Pottag, French horn in this orchestra, recalls 

that Theodore Thomas had been using tubens long before 
Mr Pottag joined the Chicago orchestra in 1907, and ever 
since, he has been playing the tuben parts when called for. 
Anton Bruckner employs eight horns and four tubens in 
each of his Seventh, Eighth and Ninth symphonies. The 
Chicago Symphony Orchestra horn quartet played the four 
tuben parts in Wagner's "Ring" for years with the Chicago 
Civic Opera Company, when Campanini was impresario. 
Other symphony orchestras using genuine tubens are the 
Philharmonic-Symphony Society of New York and the 
Philadelphia Orchestra, but in many other orchestras the 
tuben parts are given to other instruments. 

Wagner wrote a great many beautiful horn passages. One 
of the most dramatic is the famous horn call of Siegfried. 
At one place in the opera this stirring call is supposed to be 
heard in the distance. To achieve this effect, the horn player 
sounds the call backstage. Arthur Berv, first horn of the 
Philadelphia Orchestra, relates an almost tragic experience 
when he was called upon to play this well-known passage 
with the orchestra while on tour. To produce the far-away 
effect, he chose a spot just outside an exit door, permitting 
the horn music to reach the audience through this passage- 
way. Standing there ready to play the call, with just a few 
bars to wait, he was accosted by a custodian of the theater. 
The custodian told him in no uncertain terms that fire regu- 
lations or some other rule required that this door be kept 
closed. Mr Berv's frantic explanations seemed to mean noth- 
ing. Just at the point where the custodian threatened to close 
the door by force, he relented and permitted Mr Berv to 
play his call; but during the altercation the cue was missed, 
and although the call was finally given, it was a bar too late. 

[2OO ] 


After the concert Mr Berv hastened to explain to the con- 
ductor why he had come in so late, and the conductor, at 
first somewhat irritated because of the tardy horn call, broke 
out into a hearty laugh when he learned of the poor horn 
player's predicament. 

This Siegfried horn call is constantly getting into trouble. 
Once in Boston the horn player was out backstage waiting 
to play the passage, when a stagehand spotted him. Being 
new on the job and very zealous about his duty, he refused 
to permit the horn player to give his call. Remonstrating with 
the horn player, the stagehand said, "You can't play here. 
Why, can't you hear? there's an orchestra playing in there/' 
The cue came and went, but the horn call never was sounded. 

No other instrument of the brasses can equal the French 
horn in velvety tone, playing range, dynamic expression and 
variety of effects. Not only is it one of the most expressive 
solo instruments of the brasses, but it has a unique quality 
of tone which blends well with strings, woodwinds or brasses. 
When played mezzo forte, its tones melt into those of the 
strings or woodwinds, and when played forte it can give a 
good imitation of the trumpet or trombone. It can match 
tfpes with the delicate flute or with the deep-voiced, dark- 
colored bassoon. It is difficult to describe the tone quality 
of the horn to those who have never heard it. It has rare 
tonal tints which no other instrument has and which no one 
can ever forget after having once heard them. It is mellow 
but highly penetrating and expressive when played naturally, 
but when forced it becomes bold and powerful. The muted 
notes, when played naturally, produce beautiful far-away 
and echo effects, but when forced they give peculiarly 
demoniacal or terrifying effects and are often used to depict 


themes of this nature. An example of the latter occurs in 
Gounod's "Faust." The very voice of the devil comes from 
the horns as the aged Faust is inveigled into selling his soul 
for youth. These sounds from the " blasted " horns make the 
flesh creep. Terrifying are the " blasted " horn notes written 
by Wagner to signify the death of Siegfried in "Gotter- 
dammerung." No other instrument can quite imitate these 
dreadful, ghastly sounds. 

In the middle register, from Bb below the treble-clef staff 
to F top line, the French horn can be depended on for almost 
any demands. It can play extreme pianissimo and swell to 
powerful fortissimo and die back again into pianissimo with 
the most surprising smoothness. Both legato and staccato 
playing come easily for most players, as do trills, especially 
in the upper part of the scale. These are made both with the 
valves and also with the lips. Because of the close spacing 
of the open notes, lip arpeggios are very fine on the horn. 

While all other valve instruments are played with the 
right hand, the French horn is played with the left. The 
reason for this is to be found in the history of the horn. 
When Hampel discovered how to obtain an additional series 
of notes by stopping the horn, he naturally employed the 
right hand because it was more facile than the left. This very 
important job of stopping the horn was most naturally 
delegated to the right hand, leaving the left hand nothing 
to do but hold the horn. Later, when Blumel invented the 
piston and rotary valves, stopping the horn was still very 
important. In fact, when the valves were first used, the 
natural hand horn was called for as much as the valve horn. 
The player would be asked to use first the hand horn and 
then the valve horn. Even today, stopping the horn is still 



Important. There never was, therefore, a distinct break 
between purely hand-horn playing and valve-horn playing. 
The right hand continued to be used to stop the horn, and 
that is the way we find it today. Stopping is not as important 
as it used to be before good valves were manufactured and 
before satisfactory mutes were developed, and there is no 
question that greater speed could be developed in operating 
the valves with the right hand, but custom decrees that the 
valves shall be operated by the left hand. If some manu- 
facturer should get the courage to make a French horn in 
which the valves were operated by the right hand, it is ex- 
tremely doubtful if it ever would be used. The French horn 
probably will always remain the one exception to the use of 
the right hand to operate the valves. 

There is considerable confusion about stopping. About half 
the treatises dealing with the French horn say that stopping 
the horn raises the pitch, and about half say that stopping 
the horn lowers the pitch. In the less critical of these, no 
distinction is made between partial stopping, such as half- 
and three-quarter, and full-stopping. It is pretty well agreed 
among most horn players that partial stopping does flat the 
open notes. From this well-established fact, many jump to 
the conclusion that full-stopping must also flat open notes, 
only more so. Some contend that since an act carried only 
part way flattens the open notes, it is illogical to believe that 
this same act carried further should have an opposite result. 
The subject is really one without practical value, since some 
of the finest teachers and some of the finest horn players in 
the world have enjoyed great success while believing that 
when they full-stop the horn, the pitch is raised. For all 
practical purposes this is true. 


btained by full stopping 

obtained by l/ 2 s to PP in g 

G on the second line Is open, and so is E on the first line. 
Gb is obtained by half-stopping, while F is obtained by full- 
stopping. Now, the burning question is this: is F (which 
everyone admits is obtained by full-stopping) the result of 
flatting the open G one full tone or the result of raising the 
open E a half-tone? Most French horn players contend that 
the open E is raised a half-step. This view seems to obtain 
strong support from instruments having "stop valves." 
These stop valves are used in order to add tubing sufficient 
in length to flatten any note a half-step. When the horn is 
full-stopped, the whole instrument plays a half-step sharp, 
and the stop valve is used to compensate for this half-step 
sharpness, so the player can play the stopped horn in tune. 
If full-stopping the horn did not raise the pitch, it does seem 
exceedingly strange that players could succeed so well if their 
playing were based on false principles. Someday a sound 
engineer or other acoustical expert may demonstrate that 
full-stopping actually does flatten open notes, but in the 
meantime and afterward, too horn players will play just 
as beautifully, regardless of whether full-stopping flats or 
sharps the open notes. 

The French horn is considered difficult to play, and for this 
reason the mellophone is sometimes used as a substitute. 
Both are built in F, and both play the same music. Since both 
are voiced the same, one would naturally suppose they were 
the same in length, but the French horn is twice as long as 
the mellophone, the former being about 147 inches long, the 


latter being about seventy-one inches long. How then can 
they play the same music ? 
The scale of open notes in the French horn is as follows : 

A 15 16 17 1?^1 9 J2 

iS *J 1 

-ft- li i "ft* , 

=7 * 


ity a 10 

U5 3 


The C an octave below Middle C is the second note in the 
scale and not the fundamental of the horn. The diameter 
of the horn tubing is so small in proportion to its length that 
this lowest note is not used. The length of the horn is suffi- 
cient to give a fundamental an octave lower, but the tubing 
is so small that this note is not obtainable. The eleventh, 
thirteenth, seventeenth and nineteenth notes of the scale are 
badly out of tune with evenly tempered scale and are not 
used. These are shown above as solid notes, the open notes 
all being in tune and used. 
The scale of open notes on the mellophone is as follows : 

The lowest note on the mellophone is the same as the lowest 
on the French horn, but it is the fundamental, or first open 
note, of the mellophone and not the second in the scale, as 
is true on the horn. The horn plays upward from its second 
open note, while the mellophone plays upward from its first 


open note. This explains why these instruments, although 
of unequal length, play the same music. 

The respective scales of open notes also explain why the 
French horn is more difficult to play than the mellophone. 
It has been observed in several previous chapters that the 
open notes in the scale become closer together the farther 
up in the scale these notes are found. By comparing the scale 
of open notes of the French horn with that of the mellophone 
and noting particularly the difference in the spacing between 
the notes from C in the staff to C above the staff, it is clear 
that the notes on the French horn are close together, while 
on the mellophone they are comparatively far apart. In the 
octave just cited, there are only five notes on the mellophone, 
but there are nine on the horn. If there were nine bottles 
lined up on a box just a yard long, it would be more difficult 
to hit any one bottle than if there were only five bottles on 
the box. It would require a better marksman to pick out any 
one bottle among the nine than it would among the five 
bottles. That is somewhat the case in playing the French 
horn. The notes are so close together that it requires a well- 
trained embouchure, or playing lips, to reach up among 
this cluster of notes and pick out just exactly the one 
required. On the mellophone the job is comparatively 
simpler because the notes are farther apart and more easily 

Although somewhat easier to play, the mellophone can 
never be a substitute for the French horn. The quality of 
tone is not there. The larger-bore tubing of more abrupt 
conical shape lacks the richness of the small, gradually 
tapered tubing of the horn. The mouthpiece of the mello- 
phone is larger, more cup-shaped and consequently easier 



to control. The mouthpiece of the French horn is small, 
funnel-shaped and more difficult to blow. But no instrument 
can take the place of the French horn, and players are glad 
to devote themselves to the pleasant, if exacting, task of 
coaxing the velvety tone from the horn. 



The Trombone 


FOR CENTURIES THE TROMBONE was merely a big trumpet, 
and its history is inextricably bound up with the early 
history of the trumpet. In fact, its name trombone means, 
in Italian, "big trumpet." It is cylindrical in about two thirds 
of its length and conical in the other third, somewhat as the 
trumpet. Its tone is bright and noble, suggesting a trumpet 
of lower pitch. But it became an individual instrument on 
that distant day when somebody added the slide feature. 
There are many delightful stories about the origin of the 
trombone. One of the most widely related is that the Spartan 
bard, Tyrtaeus, invented it in 685 B.C. He is supposed to 
have obtained his idea from playing a trumpet with a tuning 
slide. To support the story of early invention, a trombone is 
supposed to have been found some years ago in excavations 
of the ruins of Pompeii, which was destroyed in 79 A.D. 
None of these stories has been definitely verified, and most 
authorities have ceased to regard them seriously. The best 
information obtainable is that the trombone slide principle 
was invented in northern Italy during the early part of the 
fourteenth century. 


Nobody will ever know exactly who was responsible for 
this great invention or just how it came about. It is con- 
jectured that the idea grew out of the use of a tuning slide. 
But tuning slides themselves must have been a rare device 
on a musical instrument and may have been a product of the 
same locality during the same era as that which produced 
the trombone slide. Granting that the idea arose from the 
use of a tuning slide, it is obvious how this device would sug- 
gest the slides now found on trombones. The player probably 
noticed that when he tried to tune his instrument while 
playing, the pitch was lowered when the slide was drawn out 
and was raised when the slide was pushed in. 

If this fourteenth-century player were an especially intelli- 
gent musician he probably worried much about the limita- 
tions of his instrument. He no doubt wished it were possible 
to play on his big trumpet all the notes which the wandering 
troubadour could play on his fiddle. He also, no doubt, felt 
his instrument inferior to the simple flute, with its many 
holes in the side of the body. To try to make music with a 
halfiillozen solitary notes was a disappointing task. The 
single trumpet was all right for military bugling, but as a 
musical instrumenl^t had serious shortcomings. . . __ _ 
"^"XSTthis occasion when the slide idea was born, the trump- 
eter noted that moving the slide gave additional notes. The 
half-dozen which he had become so tired of playing suddenly 
multiplied. If the slide happened to be of such a length that 
when it was drawn a note was flattened a semitone, the 
player probably observed that he could play, with the use 
of the slide, the flats of his half-dozen notes, giving him a 
scale of a dozen notes. He also clearly saw that if the slide 
were long enough to lower the tone a whole step, he could 


obtain the half-step by drawing it only halfway out, and 
then could obtain the whole step lower by pulling it out the 
entire length. 

Here was an idea which had great possibilities, one which 
was to have tremendous results for future music. To obtain 
more and more notes for bridging the gaps in the scale of the 
half-dozen open notes, the slide was made longer and longer. 
At first, though, it is believed the slide was long enough for 
four semitones. To lower the Bb tenor trombone four semi- 
tones would require a slide about fifteen inches long, since 
each of the semitones would require about three and a half 
inches to four inches. This was sufficient for bridging the 
gaps in most of the scale then used and made the scale 
chromatic from the lowest note (the second open tone) to 
the top of instrument, except for a slight gap of two semi- 
tones between the second and third open tone. The player 
could go down the scale from open F in this order: F (open), 
E, Eb, D, Db, using the four semitones made available by 
the slide; but the player could not play chromatically down 
to open Bb, because there was no way of playing C, Btj. 


This section of the trombone scale shows the open notes of the 
scale as whole notes, the numbers below denoting their order in the 
natural scale. The first of these was probably not playable on the 
early trombone. The solid notes 'show the notes playable by using 
the four positions of the slide. The X's indicate the missing notes, 
later added when the slide was lengthened. 

But why want too much ! The four semitones gave such great 
improvements over anything ever known before that it 
seemed tempting fate to ask for more, or even to think for a 



(i) Clarion in Bb, used in Florence in early fifteenth century. (2) English sackbut 
in Bb. Originated in northern Italy in fourteenth century and spread to Europe in 
the fifteenth century. (3) Modern Conn Connqueror trombone with Vocabell. 

Cross sections of the trumpet and trombone 
mouthpieces. Notice the same general shape, 
the difference between them being mostly 
size. Similar shape helps give similar tone 

The bass trombone, with rotary-valve change to F and E, bridging 
the gap in the bottom of the scale of open notes. 

The valve trombone, developed after the invention of the piston 
valve in 1815. Very little used in America. Sometimes called the 
Italian trombone because of its use in Italy. 

moment that this great slide trumpet with an additional 

-ssinijtones for every open note was not indeed perfect! 
This instrument with the slides was introduced into' Eng- 
land before the fourteenth century closed and was given the 
name of sackbut. This name was apt, being derived from 
Spanish and meaning "a pump." Henry VII was fond of band 
music, made up of crude oboes, flutes, trumpets and drums, 
and according to one account his band in 1495 contained 
four sackbuts. Henry VIII increased the number of sackbuts 
to ten, which would seem to indicate they were a kind of 
favorite of his. 

""- -As the sackbut was used more and more, this gap in the 
scale between the second and third open notes seemed to 
many musicians a hindrance to the full development of the 
musical possibilities of the instrument. Finally "perfection" 
was again reached by increasing the length of the slide about 
four inches so an additional semitone could be added. Vir- 
dung, one of the earliest European historians on musical 
instruments, wrote in 1511 that there existed at that time a 
marvelous instrument with a slide long enough to add five 
semitones to the natural, or open, scale. This extra semitone 
probably seemed at the time more than enough for all musical 
purposes, just as, in the development of the flute, the first 
three or four keys seemed all that could ever be needed. But 
these five semitones could lower the scale chromatically from 
open F only to Ct], leaving the Bt( unplayable. But, again, 
what was one semitone! It was some years later before the 
sixth semitone was provided by making the slide longer. 
This is the number of semitones possible on the modern 
tenor slide trombone, and it makes the scale chromatic from 
the second open note to the top of the instrument. 


In 1520 the Brescian school of violin making became 
famous as a source of fine violins, and, by a strange coinci- 
dence, in this same year the Hans Menschel family of Nurem- 
berg became famous as a source of fine trombones. Another 
parallel between the violin and the trombone is the fact that 
they are the two outstanding instruments which can be 
played in perfect tune. Instruments with keys and valves 
have certain inaccuracies in their scale, but the ability to 
place the finger any place on the string of the violin and to 
place the slide in any desired position on the trombone 
enables the musician to play a true and accurate scale on 
these two instruments. This parallel makes us wonder if the 
fact that fine instruments of each type were first built in 
1520 the violin in Italy, the trombone in Germany is not 
more than a coincidence. Possibly each was the flowering of 
the great musical urge which was working in both of these 
countries at this time. 

In Germany, in 1558, a composer named Kruger published 
a volume of four-part chorales, for four and six trombones, 
to be played along with the organ. No doubt some of the 
fine trombones made by the Menschel family were used by 
musicians to play these compositions. More than likely 
Kruger was inspired to write this music because of the fine 
trombones which were made available to trombone players 
by Hans and other members of this family. Composers do not 
write for instruments which do not exist, nor beyond the 
known capacity of existing instruments, but they are quick 
to employ new instruments which seem to have musical 

In Italy the creators of opera were using trombones. Two 
of the earliest, Striggio and Corteccia, used four trombones 



in 1565, along with cornets and other instruments. In France, 
Balthasarini gave a performance in 1581 of his "Ballet 
Comique de la Reine," and in the odd assortment of instru- 
ments used for instrumental accompaniment trombones were 
used, as well as cornets. Monteverde, an ingenious Italian 
composer who was quick to recognize merit in musical instru- 
ments, called for five trombones in the performance of his 
opera "Orfeo," given in 1608. These five trombones were 
divided into three voices: two altos, two tenors and one bass. 
Praetorius, writing about this time, said the trombone family 
consisted of an alto in D, three tenors in A, E and D (the 
latter an octave below the alto), and a bass in A, an octave 
below the tenor in A, 

Although Bach and Handel used the trombone occasion- 
ally, they showed no improvement in the use of this instru- 
ment over their predecessors. Bach didn't seem to care a 
great deal for the instrument and used it only sparingly, 
Handel used the trombone probably less than Bach, but 
when he did write for it, he wrote effectively. No doubt, one 
of the reasons Bach and Handel did not write more for this 
instrument was that they were so engrossed with their high 
clarion writing. They were not even content to write in the 
range of the soprano trumpet they concentrated on the 
sopranino clarion. With their attention turned toward the 
high sopranos, it is no wonder they minimized the tenor voice 
of the brass family. The most effective writing Handel ever 
did for the 'trombone was not for the tenor but for the alto 
in A, which was really a soprano in A, being of the same 
pitch as today's A trumpet. This effective writing can be 
heard as the obbligato to the voice which sings "The Trumpet 
Shall Sound," in the "Messiah." 

During the latter part of the eighteenth century, the com- 
posers of opera first learned the effective use of the trombone. 
Gluck was one of the first and one of the greatest masters of 
the instrument. Up to his time the instrument had been used 
chiefly as a volume maker to strengthen the choral or instru- 
mental ensemble in tutti passages. It doubled trumpet parts 
or played along with the voices an octave below. Gluck saw 
greater possibilities for the instrument. He liked its coloring 
and felt it could sing a song of its own. He began giving 
melodies to the trombones and scored for them in three parts. 
This three-part harmony was tied in with the lower voices 
by giving the fourth note of the chord to the bass instruments 
or having them double the third-trombone part an octave 
lower. This finely knit style of writing, rich in colorings and 
varied in pattern, set the style of trombone writing from that 
time until today. Opera composers adopted this style almost 
immediately, because they saw its superiority over what had 
been done before, but it was not until Beethoven's time that 
it was used in the symphony orchestra. This form of part 
writing for the trombone section first made its appearance in 
symphony, in the Finale of Beethoven's famous Fifth. 

Gluck was a German but wrote in France during most of 
his life. Gossec, however, is given credit for introducing 
trombones into French opera, when he used them in 1770, in 
his opera "Sabina." Gluck and other German composers 
usually called for three trombones the alto, tenor and bass 
but French and Italian composers ordinarily used only 
the alto and tenor. It was not until the latter part of the 
nineteenth century that it became fairly general for com- 
posers in all nations to use three trombones, but they were 
not the alto, tenor and bass of Gluck and other Germans, but 



two tenors and a bass. The small alto fell into disfavor, and 
a second tenor trombone took the top notes. The range of 
the trombone trio was extended into the bass, and the bass 
member of the family became more important. 

In England the adoption of the sackbut in the orchestra 
was delayed because of its association with the cornetto and 
its consequent decline along with this instrument. The cor- 
netto and sackbut were usually used together in English 
bands and church choirs, one cornetto as treble to three 
sackbuts as alto and tenors. When the cornetto lost favor 
after 1650, the sackbut also was neglected. For nearly a cen- 
tury there was little call for the sackbut, and it was late in 
the eighteenth century before some of its old popularity 
returned. It came back also with a new name trombone. 

During the period when the trombone was staging a come- 
back, it seems the English people went in for slide instru- 
ments in a big way. They not only became interested in the 
slide trombone but also in the slide trumpet, so much so 
that while other countries in Europe were welcoming the 
valve trumpet, England was too engrossed in the slide 
principle to get interested in the newly invented valves. The 
enthusiasm for the slide principle, not only in trombones but 
also in trumpets, delayed the use of the valve trumpet in 
England until well into the nineteenth century. 

While the slide trumpet has been more fully discussed in 
the story of the trumpet, it is interesting here because of its 
slide principle. Far from being like the slide trombone, the 
slide trumpet is interesting in the discussion of the slide 
trombone chiefly because it is so different. Some have con- 
fused this instrument as being in reality a treble trombone. 
This is far from true. With the passing of the cornetto, which 

took treble parts with the trombone, the treble trombone did 
have a vogue. The "tromba di tirarsi" of Bach and Handel 
is a genuine trombone built in treble A or Bb, an octave 
above the corresponding tenors in these keys. Such treble 
trombones had a short period of popularity, but it passed 
after Bach and Handel. The slide trumpet, however, was 
different from this treble slide trombone. The slide on the 
trombone was pushed forward, or away from the player, in 
order to lower its tone. The slide on the slide trumpet was 
operated in exactly the reverse manner. To lower the tone, 
the slide was drawn toward the player. The slide on the 
trombone was out in front of the instrument, while the slide 
on the trumpet was placed toward the back of the instru- 
ment, somewhat as the first valve slide is now placed on the 
modern trumpet. The slide of the trombone can be extended 
to six positions beyond the closed position and can lower the 
pitch of the instrument six semitones. The slide on the slide 
trumpet was short and had only two positions besides the 
closed position and could lower the pitch of the instrument 
only two semitones. 

This latter difference is the important difference between 
the trombone and slide trumpet. The two positions of the 
slide were used not so much as the trombone slide was used 
but more as the valves are now used. Instead of the second 
valve with its half-step of tubing opened by pushing down 
the valve, the same length of tubing was added to the open 
notes by pulling the slide out halfway. To lower the open 
note a full tone, the slide was pulled all the way out. In this 
manner a semitone or a full tone could be added to all open 
notes. Obviously, such an instrument was not chromatic, for 
there were gaps between the second and third, third and 


fourth, and fifth and sixth open notes which could not be 
completely bridged chromatically. 

In Germany, France, Italy and England the trombone 
was used in early opera, but it was not admitted to the 
symphony orchestra until later. Bach and Handel availed 
themselves of the resources of the trombone in their operas 
and oratorios, but not in their symphonic compositions. 
Haydn used three trombones in his great "Creations," but 
refused to admit the trombone in his earlier symphonies. 
Mozart experimented successfully with the trombone in 
some stirring passages in "Don Giovanni" and apparently 
felt more at home with the trombone in his "Magic Flute," 
written about four years later. But in spite of Mozart's fre- 
quent and effective use of the trombone in his operas, he did 
not use trombones at all in his three greatest symphonies: 
the El? Major, the B Minor, and the C, or "Jupiter." 

Beethoven was openly fond of the trombone; so much so 
that when he was buried in 1827, two of his "Equali" 
compositions for four trombones were played at his grave. 
Yet Beethoven did not use trombones nearly as often as 
might be expected. Apparently he was too bound by the 
customs of the times, which permitted the trombone to show 
what it could do in the opera but confined it to a back seat 
in the symphony. He ignored the trombone in his first four 
symphonies, written from 1800 to 1805. Finally, in 1808, the 
trombone got its first chance in symphonic music. In his 
great Fifth Symphony Beethoven is forced to call upon the 

Developing this marvelous musical structure, theme fol- 
lowing theme, Beethoven seems not to need trombones. 
Through three movements he gives utterance to the greatest 

of his musical thoughts without once bringing in trombones. 
But toward the close of the third movement, after a pro- 
longed pianissimo of strings and tympani, when the air is 
supercharged with electric expectancy, suddenly there is a 
twitch of the strings, a change of the rhythm, and the theme 
of triumphant joy for the fourth movement is stated in one 
of the most remarkable passages in all music. Then appar- 
ently Beethoven seemed to realize the inadequacy of all other 
instruments or combinations of instruments to express the 
exultant joy which surged in his thoughts and soul. This 
great emotion broke through conventional form and ex- 
pressed itself in noble, sustained chords from the trombones. 
What a triumphal entry for any instrument into the sym- 
phony! Although few instruments have been shut out from 
the symphony so long, no other instrument has ever had 
such a triumphal entry into these exclusive ranks. After 
this sensational debut, it is no wonder the trombone ever 
afterward found its place in the symphony secure. 

Incidentally, two other novices were admitted into the 
symphony at the same time: the piccolo and the contra- 
bassoon. When Beethoven's soul surged forward with its 
expression of hope and joy, he seemed to cast aside precedent 
and to think only of putting into musical form this great 
inspired feeling. Whatever musical coloring was needed, he 
appropriated. The trombones definitely characterized this 
supreme utterance, but the contrabassoon added certain 
purple tones at the bottom of the tonal picture, while the 
piccolo was needed to put the bright, silvery sharpness at 
the top. It was a moment in which the fetters of Beethoven's 
soul were broken, and the trombone, contrabassoon and 
piccolo for the first time found their own freedom. 


Beethoven used two trombones (and a piccolo) in some 
descriptive passages in his Sixth Symphony, but strangely 
enough, the trombone was left entirely out of his Seventh 
and Eighth. After he used trombones in his Fifth and Sixth 
symphonies, he seemed to drop them entirely. It was not 
until sixteen years after the tremendous trombone music of 
the Fifth that he again used trombones in his symphonic 
compositions. This was in his Ninth Symphony, written in 
1824. Probably the reason for using trombones in this or- 
chestration was that he was writing for an extra-large 
orchestra. Or, more probably, he felt this "Ode to Joy" 
theme required the peculiar talents of trombones. In any 
event, three trombones were used. 

The cause for Beethoven's neglect of trombones for six- 
teen years may be hinted at in a famous statement made 
later by Mendelssohn, setting forth his own regard for the 
trombone. He said he regarded the trombone as too solemn 
and noble an instrument to be used except on very special 
occasions. Perhaps Beethoven felt likewise. Perhaps during 
the writing of his first four symphonies he failed to find an 
occasion important enough to use the trombones, and then 
had to wait sixteen years after that sublime outburst in the 
Fifth before another occasion came along in the Ninth 

Schubert, Schumann and other composers who followed 
Beethoven accepted the trombone as a regular member of 
the symphony orchestra. Berlioz in his " Requiem " called 
upon the trombones to perform the stupendous task of 
simulating Gabriel's trumpet blast on the Day of Judgment. 
He outdid all other composers by specifying sixteen trom- 
bones! Not only that: he demanded that the trombones 

produce certain low notes which were not ordinarily played. 
He wrote pedal notes A, G# and F, below the lowest open 
note. Realizing the consternation of the trombone players 
when they came to these notes, he anticipated their objec- 
tions by writing in the margin: "These notes are in the 
instrument and the players must get them out." As usual, 
Berlioz understood the resources of the instrument even 
better than the players themselves, and this example has 
served to show all the musicians who followed him that these 
are legitimate notes on the trombone at least on the trom- 
bone of larger bore. 

Wagner usually wrote for three trombones, but in his 
"Ring" operas he wrote for four. Wagner was a past master 
in writing for trombones, as he was with all brass instru- 
ments, but he often seems to forget the usual duties of the 
first trombone and writes extremely low notes for it. To play 
these parts the first trombonist must have the thumb valve, 
or F attachment, on his instrument, an attachment usually 
used on only the bass trombone. Bellini does the same thing 
in his "La Somnambula," where he writes an Eb below bass 
clef for first trombone, a note which can only be played with 
the F attachment. The functions of this attachment will be 
more fully explained later. 

Shortly after the invention of valves by Blumel in 1815, 
valves were added to the trombone. This made it like a big 
valve trumpet in everything but appearance. It still retained 
the trombone appearance. But while the tone quality re- 
mained substantially the same, its style of performance was 
changed. It lost its perfection of scale, inheriting all the evils 
of faulty intonation found in valve trumpets, only more so. 
It lost the characteristic glissando effects which were due 

F 220 ] 


tol the slides. It failed to achieve acceptance generally because 
itj was considered inferior to the slide trombone. Strangely 
enough, about the only place where the valve trombone did 
njieet with favor was in Italy, the birthplace of the slide trom- 
blone. The valve trombone has become so definitely identi- 
f|ed with Italy that in America the valve trombone is often 
known as the Italian trombone. 

I Above it has been noted that Beethoven introduced the 
|:rombone into the symphony upon a very special occasion 
and waited sixteen years before he found another musical 
idea which seemed to warrant the services of this great and 
noble instrument. Mendelssohn frankly stated that he felt 
the trombone was too sacred to be used except upon the 
rarest occasions. Berlioz likewise held the trombone in great 
reverence. He protests against the indiscriminate use of the 
trombone, saying this "is to impoverish, to degrade a mag- 
nificent individuality; it is to make a hero into a slave and 
a buffoon." 

Probably one of the finest, if not the finest, character- 
izations of the trombone is to be found in Berlioz* monu- 
mental work, Modern Instrumentation and Orchestration. 
In it he says : 

"The trombone is in my opinion the true chief of 
that race of wind instruments which I have designated as 
epic instruments. It possesses, in an eminent degree, both 
nobleness and grandeur; it has all the deep and powerful 
accents of high musical poetry from the religious accent, 
calm and imposing, to the wild clamors of the orgy. It 
depends on the composer to make it by turn chant like a 
choir of priests; threaten, lament, ring a funeral knell, 
raise a hymn of glory, break forth into frantic cries, or 


sound its dread flourish to awaken the dead or to doc|<m 
the living. ... IS 

. 71 

"The character of tone in trombones varies accordimig 
to the degree of loudness with which their sound is emitted! 
In a fortissimo it is menacing and formidable. . . . Suc|h 
is the terrific scale in D minor, upon which Gluck hats 
founded the chorus of furies in the second act of hi|s 
'Iphigenia in Tauride/ ... In simple forte, trombone^ 
have an expression of heroic pomp, of majesty, of lofti-|- 
ness. . . . Then they acquire with enormously increased 
grandeur the expression of trumpets; they no longer 
menace, they proclaim; they chant, instead of roar. . . . 
In mezzo forte in the medium, in unison or in harmony 
with a slow movement, trombones assume a religious 
character. Mozart, in his chorus of the priests of Isis, in 
the 'Magic Flute/ has produced admirable models of the 
manner of giving these instruments a sacerdotal voice and 
attribute. . . . The pianissimo of trombones applied to 
harmonies belonging to the minor mode, is gloomy, 
lugubrious I had almost said hideous. If, particularly, 
the chords be brief and broken by rests, it has the effect 
of hearing some strange monsters giving utterance, in dim 
shadow, to howls of ill-suppressed rage." 

Since the trombone, as Berlioz says, belongs to that race 
of instruments called "epic," we would naturally expect it to 
move through music at a majestic pace, and so it does, some- 
times because of serious limitations. There are many musical 
figures which are impossible on the trombone. Trills, except 
some lip trills in the upper register, are entirely impractical; 
and even those which are practical are seldom played because 

[ 222 ] 


they require lip training which is beyond most players. Pure 
legato is also impossible in some parts of the scale because of 
the limitations of the slide movement. Only if the notes 
played are based on the same fundamental can pure legato 
be accomplished. To play several notes in the lower positions 
of the slide and then have to bring the slide to closed position 
for the other notes, breaks the legato quality. It is also obvi- 
ous that the rapid repetition of certain notes is impossible 
or impractical, where the slide must be moved from closed 
or first position to the sixth or seventh positions, or vice 

But there is one technical resource which the trombone 
has to the highest degree this is glissando. With the slide 
it is possible to "smear" the seven semitones encompassed 
within the limits of the slide into one continuous, integral 
note of ascending or descending pitch. A less exaggerated 
form is the slur, which the trombone can do beautifully. 
While the slur is used legitimately in all music, the glissando 
is chiefly associated with the Negro minstrel show or hot 

It is this glissando or "smear" effect which makes the 
trombone the Dr Jekyll and Mr Hyde of the musical world. 
Contrary to popular opinion, the distinction of being the 
jazziest instrument in the orchestra or band goes, not to the 
saxophone, but to the trombone. Known and admitted to 
be the most noble of all musical instruments, it is at the same 
time the jazziest of the whole lot. It depends upon the player. 
When played as a "slip horn," nothing can equal it as a 
means of expressing jazz. "Hot licks" come natural to this 
instrument. Its slide gives it unmatched resources for slur- 
ring, portamento and glissando, in which are all the sug- 


gestiveness, smirk and sensuousness for which jazz is noted. 
Modern muting and "wow-wowing" were early used on the 
"slip horn/' It can moan, it can scream, it can seduce, it can 
laugh like a hyena. It is the most gifted of all instruments 
called upon to give expression to the worst type of jazz 
mania. But when Mr Hyde steps forward, all is noble, sacred, 
grand. Whenever composers wish to depict religious solem- 
nity, or martial glory, they call upon the noble trombone. 
Wagner used it often in his allegorical and religious "Nibe- 
lungen." When Berlioz wished to portray Gabriel's awful 
trumpet announcing the Day of Judgment in "Requiem," 
he called for sixteen trombones ! The trombone is like a great 
and gifted person whose talents may be turned to good or to 
bad, but in either case great achievement is assured through 
unusual capabilities. 

v Few laymen understand the mechanical operation of the 
trombone and fewer still its musical principle. Like the rube 
at the circus, the average person knows the trombone player 
doesn't swallow the slide, but beyond that he seldom worries. 
The heart of the trombone is of course the slide. The mouth- 
piece goes into one side of the slide, the other side connecting 
with the bell. These two straight slides are joined at the far 
end by a U-shaped piece of tubing called the outside slide, 
which telescopes over the two straight inside slides. It is this 
outside slide which is operated by the right hand, which 
pulls it in and pushes it out. When the slide is in, only the 
natural scale of the instrument is sounded, just as on the 
bugle or other simple horn. But when the slide is extended 
into each of the lower six positions, six semitones are added 
to each open note, as has been described earlier in this chap- 
ter in discussing how the trombone came to be developed, 


No wonder the musical principle of the trombone is con- 
fusing. There seems little rhyme or reason to the instrument 
when the uninitiated watches it being played. The player 
may be performing a well-known tune, and the notes may 
descend gradually and slowly down and down, until the 
observer begins to anticipate the further lengthening of the 
slides as he waits for the descending notes. As the spectator 
waits for the slide to descend for the next lower note, he is 
puzzled to see the player pull the slide clear in to closed 
position just where, by all the laws of relative position, 
it should not be. Vice versa, when the observer expects the 
slide to be drawn in for a higher note, he is nonplused to see 
the player shoot the slide out to full length. He knows that 
by all the laws of relative position the note cannot be out 
there, but it is. It is a puzzler. 

Everything becomes plain when a little explanation of the 
musical principles of the trombone is given. In closed or first 
position, when no part of the outside slide is effective except 
the curved bow at the end, the natural scale of open tones is 
sounded by changing lip and blowing pressure, as follows: 


The gaps in this natural scale are bridged by using the slide, 
as already explained, but the perversity of the slide can be 
understood only by charting some of the notes made by 
using the slide. Below is reproduced a section of the trom- 


bone scale. Open notes are shown by half-notes; notes made 
by using the slide are shown by quarter-notes. Under each 
note is shown the position of the slide, I being closed, 2 being 
the first extension of the slide, or second position, 3 being the 
next, etc. 

1 2341 23451 2 3 '4 '5 67 1 

Imagine a slow passage descending from D above the bass 
staff: the slide is dropped to fourth position for B, and just 
when you think the player is going to drop the slide to a still 
lower position for Bb, you are fooled, the player pulling the 
slide in to first or closed position, which you feel should by 
rights give a higher note! Then A is obtained by extending 
the slide about four inches to second position, G is obtained 
by extending the slide an additional seven or eight inches to 
fourth position; but just when you think you have everything 
figured out and when you expect the player to extend the 
slide seven or eight inches still further to sixth position for F, 
lo and behold, the player pulls the slide in, which you swear 
should give a higher-pitched note: but it doesn't, it gives a 
lower note! The interval between F and Bb in the staff gives 
a clear notion of just how the open note (F) is sounded and 
how the lower semitones are obtained successively by ex- 
tending the slide first into second position, then into third 
position, and so on, to obtain the chromatic intervals between 
the gaps in the open notes of the natural scale. When the 
slide is extended to the seventh and final position and the 
player cannot descend any farther by using the slides, the 


Showing the positions of the slides as used to bridge the gaps in 
the open-note scale. When the slides are placed in these positions, 
a chromatic scale is possible except in the bottom of the range. 
These positions are only approximate, correct intonation requiring 
certain variations. 

next chromatic step is found to be another open note (Bb), 
played in first or closed position. 

In this manner it is possible to descend chromatically to 
E below the staff, but this note is obtained by using the 
seventh and last position. Between this note and the next 
open note (Bb) is a gap of five semitones. On the regular 
tenor trombone, these notes cannot be played, but on a tenor 
trombone with a valve attachment added, generally known 
as a bass trombone, these notes can be played and the gap 
bridged chromatically. This is done by operating a valve 
with the thumb of the left hand which adds a length of tubing 
capable of throwing the trombone from Bb into F below. 

Gap of five semitones, not playable on the tenor trombone in 
Bb. These notes are played by putting trombone into F or E by 
means of valve operated by thumb of left hand. Such a trombone 
is usually called a bass trombone. 


I Flayed on Trombone in Fi 

These notes, with slide positions marked above, are based upon 
open F of trombone when lowered to key of F by operating thumb 
valve. They serve to bridge the gap found on the Bb tenor trom- 

5 23 4 5 61 

If bass trombone is put into key of E by thumb valve, the E 
is open and the other notes down to Btj are played by the slides 
in positions as marked above. 



This length of tubing, which is added by opening the valve, 
really adds five or six semitones to the total length of trom- 
bone tubing. When in F, the F below the staff is open or 
first position. By keeping the trombone in F, it is possible to 
extend the slide first into second, then into third, etc., posi- 
tions to play E, Eb, D, Db, C, B. When the player comes to 
B, he has run out of slide again, but by closing the F valve 
and putting the trombone back into the key of Bb, the player 
finds that the next chromatic note is open Bb, played with 
the slide closed. Then, by extending the slides, this funda- 
mental can be added to and some so-called pedal notes se- 
cured. These are difficult to obtain, but are used as far down 
as the seventh position. This gives a chromatic scale from 
low E, below the fourth line under the bass staff, to D on the 
fifth line above the bass staff, a total chromatic range of 
nearly four octaves. 

Adolphe Sax was the inventor of this valve for the trom- 
bone, according to Berlioz. Before this ingenious device, 
composers could bridge this gap only by using trombones of 
different pitches, selecting the notes from one to bridge this 
gap on the other. Although today such a trombone with 
thumb valve is called a bass trombone, when it was invented 
it was still called a tenor trombone, which in reality it is. 
The bore is large, it is true, but often is not any larger than 
some large-bore tenor trombones which do not have this 
device. The chief reason for calling such a trombone a tenor 


trombone was that in those days they used a true bass trom- 
bone which played an octave below the tenor. The name 
tenor had to be preserved for the tenor with the valve attach- 
ment in order to avoid confusion with the real bass trom- 

This bass trombone was a monstrous affair with double 
tubing for slides. On the tenor trombone, when the slide is 
extended about four inches to a lower position, eight inches 
of tubing is actually added, twice the extension. On the bass 
trombone, for every inch the slide was extended four inches 
of tubing was added to the trombone. The slides were so long 
that a handle had to be used for the lower positions. Although 
usually pitched in Eb, the same as the present Eb bass tuba, 
there were some made in BBb, the same pitch as the present 
contrabass tuba in Bb. Wagner calls for the bass trombone in 
some of his compositions, but since there are few such in- 
struments being played today, these notes are either played 
on the contrabass tuba or played on the tenor trombone an 
octave higher. 

In the regular dance and radio bands, trombones of me- 
dium bore are used, the tone being brilliant and cutting. 
Trombones in the symphony orchestra are of large bore, 
the tone being correspondingly bigger and broader. The first 
trombonist uses a tenor, while the second and third trom- 
bonists usually use the bass trombone with thumb valve for 
throwing the instrument into F or Eb. Usually both sides 
of the slide are of the same bore size, but in some models 
the side nearer the bell is larger; these are called duo-bore 
trombones. The former have a brighter, more solid quality 
and are generally found better for recording and radio work. 
The duo-bore trombones have a big, round tone but seem 



to sacrifice some of the brilliance usually associated with 
the true trombone quality. Control men in recording and 
radio studios find the straight-bore trombones better for 
recording and radio work than the duo-bore, because the 
latter have an uneven and jumpy quality, technically known 
as "spreading," while the former "cut through " with sharper 
and more solid quality. 


The Tubas 


IN 1590 THE French churchman Guillaume of Auxerre in- 
vented the serpent, a queer-looking wind instrument 
about eight feet long. Its tubing was made in a shape suggest- 
ing a squirming snake which had been struck with a stick. 
For about two hundred years it flourished as an important 
bass instrument, but now it is chiefly known for its many 
and varied progeny. Among these are the ophicleides, a 
family of six; the saxhorns, a family of eight; the saxtrombas, 
a family of eight; the tubas, a family of nine; and the much- 
maligned saxophones, which have now grown to a family of 

Too much credit cannot be given Guillaume for his in- 
vention, because the serpent is little more than a bass mem- 
ber of the large family of cornettos, or zinken. These instru- 
ments put in an appearance in Europe in the fourteenth 
century. In England they were called cornettos and were 
built in three keys. The little treble cornetto in F was only 
about eighteen inches long and had a thin, weak tone. 
Another was the cornetto in C, about two feet long* The 



third was the great cornetto in G, approximately three feet 
long. In Germany these same instruments were known as 
zinken and were built in several keys, one of them being a 
high soprano in D which was only a little over a foot long. 
It is not definitely known how many different members there 
were in this original family of instruments, but there un- 
doubtedly were quite a number. 

The cornettos and zinken were the "poor white trash" 
among musical instruments. Not only were they cheaply 
made, but they were noted for their poverty of musical 
qualities. Constructed of wood and covered with leather, 
their tone was colorless, coarse and windy. Anyone with a 
pocketknife, a pot of glue and a thin skiver of leather could 
make one of these instruments. Two sides of the tube were 
whittled out and stuck together with glue. Then the tube 
was covered with leather to strengthen the thin wood. After 
this, holes were bored in the side of the tube and a cup- 
shaped mouthpiece was turned out of a piece of wood. The 
instrument was then complete. Making these ancient instru- 
ments was something like making the cigar-box fiddle or 
the slippery-elm whistle of today. 

Nevertheless, these instruments became the most popular 
in Europe. In the fifteenth and sixteenth centuries they were 
heard in military bands and church choirs and were rated 
as the most important wind instruments of their time. Their 
number and variety multiplied. They swept over Europe 
somewhat as their famous offspring, the saxophones, later 
swept over America in the 19205. Guillaume, no doubt, heard 
them from morning to night, not only outside his church but 
also in his own choir loft. Apparently Guillaume shared the 
enthusiasm of his contemporaries for these instruments. He 

decided the world would be further blessed if a larger and 
better zinke or cornetto were invented, and accordingly 
he brought forth the serpent. 

Originally it was a conical tube about eight feet long with 
six finger holes and was played with a wooden cup-shaped 
mouthpiece. Later, keys were added and the mouthpipe 
and mouthpiece were made of metal. Its pitch was two 
octaves below Middle C or thereabout, and it furnished a 
deep voice for the military bands and for the church choirs. 
It found its place in musical circles much as might a bass 
singer in a college glee club which was without adequate 
bass voices. The only important wind bass at the time was 
the bassoon, and its reedy quality of tone did not seem to 
strike theTancy of the populace. The serpent, therefore, was 
looked upon as a much-needed addition to wind instruments, 
a great boon to music. 

Fifty years before Guillaume invented the serpent, another 
churchman, Afranio of Ferrara, Italy, had invented the shape 
of the bassoon. He built it so its tubing was doubled back 
upon itself in parallel lines, which shape earned for the 
instrument the name of fagotto, or bundle of fagots. Guil- 
laume apparently did not think well of his brother church- 
man's design, for the serpent was curved into a fantastic 
shape which resembled a reptile. It has seemed odd to some 
people that Guillaume, a divine, would make an instrument 
in the shape of the serpent, symbol of evil. Possibly it did 
not occur to Guillaume that he was giving the shape of a 
serpent to his instrument. Possibly he took his design from 
the fifteenth-century painting, "Angel with the Trumpet." 
In this painting, the trumpet is shown bent in the shape of 
a reversed letter S. Perhaps Guillaume was simply trying to 


All instruments are of " backfiring' 1 variety, the bell extending 
backward over the shoulder and directing the sound to the rear. 
Useful in marching bands which played at the head of a moving 
column of men, but abandoned in favor of the bell-front instru- 
ment, which is better in concert work. 


carry this idea into the bass instrument and evolved his 
serpent shape. In any event., he produced an instrument 
whose shape has caused wonder and amazement for three 
hundred years. 

Although Afranio's bassoon design lacked a certain amount 
of showmanship which Guillaume's serpent design possessed, 
the former proved the more practical and was eventually 
emulated by the serpent. After remaining curved like a snake 
for two centuries, the serpent finally changed to the bassoon 
shape along toward the close of the eighteenth century. Its 
eight feet of tubing was placed in two parallel lines, and it 
became known as the serpent horn, or military serpent. 
Although usually built in C or D, two octaves below Middle 
C, sometimes it was built long enough to play down to A 
or G. This was getting down into the bass territory of the 
modern bass tuba. In the beginning of the nineteenth cen- 
tury there appeared the metal serpent horn, known more 
generally as the serpentcleide, or ophicleide, both of which 
mean "keyed serpent." The first of these had seven keys, 
but later the number increased to eleven. It usually was 
pitched in B, two octaves and a semitone below Middle C. 

Blood began to tell in this new tribe of instruments, for 
the number of ophicleides increased until there were six or 
seven. Besides the bass in B, there were also basses in C and 
Bb- Then contrabasses in F and Eb were created. There were 
smaller ophicleides, too, for we find altos in F and Eb. The 
zinken and cornettos had long since passed from the scene, 
but their ability to propagate had been successfully carried 
in the blood stream of the serpent and passed on to the 
ophicleides. It is possible that had Guillaume failed to invent 
the serpent, the youngest and largest of the zinken, this race 


of musical instruments would have become extinct in the 
seventeenth century. All of the smaller members of the family 
did become extinct, but this race of musical instruments was 
saved from oblivion through the virility of the large serpent. 
From the seed of the serpent grew up the family of ophi- 
cleides, and dynasty after dynasty of related instruments 
grew up, only to give way to still another. Today we have 
zinken progeny in the form of tubas and saxophones. 

Besides the popular acclaim accorded the original zinken, 
these early instruments found some favor with composers 
of serious music. The early Italian masters called for zinken 
in some of their operas. After the big bass zinke, the serpent, 
had served its apprenticeship in the military band and church 
choir, Handel scored for it in his "Water Music" and his 
"Fireworks Music/' Mendelssohn wrote parts for the serpent 
in his "St Paul" and wrote freely for the later form of the 
serpent, known as the ophicleide. Schumann used it in his 
cantata, "Paradise and Peri." Even Beethoven wrote for it, 
no doubt because there was no other bass suitable for his 
needs. But Wagner was faced with no such compulsion, for 
there were several cup-mouthpiece basses available in his 
time; yet he wrote for the ophicleide in some of his early 
scores. It appeared in Wagner's scores for the last time in his 
"Rienzi," first produced in 1842. Meyerbeer, who died in 
1864, used the ophicleide to the complete exclusion of the 
bass tuba. William Wallace in his opera, "Love's Triumph," 
written in 1862, provided parts for the bass tuba, whereas 
in all previous scores he had given the bass parts to the 

During the first half of the nineteenth century the ophi- 
cleide held a place of some prominence in music, but the 



eventual doom of the ophlcleide was sounded early in the 
century. In 1815 Blumel invented his piston valve, and 
although the keyed serpent continued on for another genera- 
tion, it was only a question of time until it was supplanted 
by the valve bass. In 1815 the ophicleide was "going strong/* 
It was probably at the height of its popularity. It was used 
in opera and in the church choir. It was a popular instrument 
in the military bands. Ophicleides were used in the bands of 
both French and English on the field of Waterloo. But the 
year which saw Napoleon go down to defeat was the fatal 
year in which sentence was also pronounced upon the ophi- 
cleide by BlumePs more efficient piston valve. 

When the piston valve was first invented, few people 
realized that the day of the ophicleide was over. Twenty years 
after Blumel made his first model, there were important 
musicians who felt that the keyed serpent was still the last 
word in basses. The beginnings of the valve bass were not 
any too impressive. Naturally the soprano and alto instru- 
ments were the first to be given the advantage of the new 
valves. It was not until 1828 that Wilhelm Wieprecht, master 
of bands of the King of Prussia, produced his family of 
valve instruments. This family was not quite complete, but 
it included a small Eb cornet, an Eb trumpet, a Bb tenor 
and a Bb baritone. Later, it seems, the bore of the baritone 
was made larger, and it became known in England as the 
euphonium and in other countries as a bass tuba. This in- 
strument was pitched in Bb ? the same as our present baritone 
and euphonium tubas, but the bore was somewhat larger 
than our present euphoniums, and the instrument was gener- 
ally referred to as the bass tuba. Bass tubas were built in 
other keys also, the F bass tuba being a popular one. Of the 


latter there seems to have been two varieties, both built in 
F, but one larger in bore than the other. Berlioz refers to one 
as the bombardon and to the other as the bass tuba. 

By the middle of the century there were quite a number of 
bass horns of various kinds and types. In Berlioz 5 famous 
book on instruments, published in 1848, no less than seven 
basses of various kinds then in use at least in France are 
described. The ophicleides were still hanging on, for Berlioz 
calls attention to the bass in B, as well as the contrabass 
ophicleides in F and Et. Then he takes up the newer bom- 
bardon, built in F and equipped with three valves. Similar 
to the bombardon, but built in larger bore and equipped with 
five valves, is the German bass tuba, a creation of Wieprecht. 
Although Berlioz praises the great sonorousness and low 
compass of the bass tuba in F, he hastens to mention that 
Adolphe Sax was at that time building a bass tuba in Efc>, 
a whole tone lower. The old wooden serpent Berlioz mentions 
rather casually, with some caustic remarks about its "bar- 
barous" tone quality. He also mentions the "Russian 
bassoon," an instrument even inferior to the serpent, and 
suggests that the art of music would not suffer in the slightest 
if it should disappear. 

With such an array of bass instruments to select from, 
composers, directors and musicians must have been in a 
constant quandary. Times certainly had changed from the 
day when the only wind basses available were the bassoon 
and bass trombone. The serpent was welcomed with open 
arms when it appeared on the scene, but the world was soon 
to find out that it brought not so much a gift of music as of 
fecundity. In Berlioz' day, the old serpent was still hanging 
on desperately but was rapidly being suffocated by its many 



offspring which were growing up around it. Before 1590 there 
was only the feeble low voice of the bassoon, with an occa- 
sional low growl from the bass trombone, but before the 
middle of the nineteenth century there was a veritable babel 
of bass voices, a babel of the children of the serpent. It was 
time somebody made a housecleaning, and Adolphe Sax 
elected himself the man of the hour. 

Adolphe Sax is familiarly known as inventor of the saxo- 
phone. He and his father, C. J. Sax, were instrument makers 
in Paris. These two men contributed many improvements to 
musical instruments, refining their response and intonation 
and inventing ingenious mechanisms which made them 
easier to play. Fresh from the triumphs attending his in- 
venting and patenting the family of saxophones in 1840, 
Adolphe saw the great need of bringing order out of chaos 
among the cup-mouthpiece instruments, especially those of 
lower voice. The result of this work was his family of sax- 
horns, brought out in 1842. Owing to the lack of uniform 
terminology in Sax's time, it is rather difficult to describe 
this family, for sopranos were sometimes called altos, altos 
were often referred to as tenors, tenors were called baritones, 
and one is never sure whether a baritone was a euphonium 
or a bass. Six years after their invention, Berlioz discusses 
eleven different members of the saxhorn family, from very 
high sopranos to low double-basses in Eb and drone double- 
basses in BBb. But put into familiar terms which have come 
to be generally accepted today, this family of saxhorns 
consisted of a small Eb soprano, similar to our high Eb 
cornet; a Bb soprano, similar to our Bb cornet; an Eb alto; 
a Bb tenor; a Bb bass; an Eb low bass; and a contrabass in 


This logical organization of the cup-mouthpiece instru- 
ments was accepted more or less quickly, not only in France, 
but also in Germany, where Wieprecht had already made 
progress along this same line. England and Italy fell in line 
later, and finally America has come to use this classification 
of the various cup-mouthpiece voices, separated by intervals 
of fifths and fourths. With the invention of the saxhorn 
family, the miscellaneous serpents, ophicleides, bombardons 
and other odd tenor, baritone and bass instruments faded 
out of the picture. 

As Haydn sounded the death knell to the rabble of in- 
struments through his orchestrating, so Adolphe Sax silenced 
the Tower of Babel through his instrument making. The day 
of Haydn had been preceded by a lot of haphazard inventing 
of instruments of all sizes, kinds and descriptions. When 
Haydn appeared he selected with great insight those instru- 
ments worthy of a place in the orchestra and sent the rest 
of the rabble to oblivion, from which not a single instrument 
has staged a successful comeback. A century later Sax found 
a similar condition. The seed of the serpent had spread and 
sprouted and had produced a confusion of voices which 
irked the systematic, straight-thinking Sax. He looked them 
all over, found them all wanting in some aspect, and then 
substituted a new family of instruments, made up of a com- 
bination of the best of them all. Having salvaged the best 
from the lot, he saw no need of saving the remnants, and so 
he, like Haydn, dispatched the motley crew into oblivion. 

Besides organizing the voices of the cup-mouthpiece in- 
struments into a logical and limited number of instruments, 
Sax also made a contribution in new tonal color. The saxhorns 
characterized by a conical bore of wide taper and a 



mouthpiece which is deep and bell-shaped. This construction 
gives to the whole family a distinctive tonal color which is 
round, mellow and on the dark side of the tonal spectrum. 
In America we hardly know what this tonal color is like, 
for our instruments have a straighter bore and are more 
brilliant in color; but the tapered bore is still much used in 
Italy and also in the military bands of France and England* 
The effect of a lot of these saxhorns playing together sug- 
gests a great pipe organ, because the tone is so full and 

AH the smaller instruments up to and including the alto 
were played in a horizontal position, as is the familiar cornet 
today, while the larger instruments were played in a vertical 
position, as is our bass or baritone. The Eb bass and the BB[? 
contrabass were also made in a circular form, encircling the 
body and resting on the left shoulder. These were called 
helicons, a name originally given by Ptolemy to a geometrical 
drawing illustrating his idea of the eternal ratios existing 
between tone and color throughout the universe. Often there 
were two Bb sopranos, both in the same pitch but differing 
in the size of bore. The smaller bore was the soprano, while 
the larger and more tapered bore was the mezzo-soprano. 
They both sang in the same voice, but the larger and more 
tapered bore instrument had a bigger, more mellow tone. 
This mezzo-soprano has come down to us in the form of the 
fluegelhorn. The bore of the Eb alto was pretty well standard- 
ized, but the bore of the Bb instruments pitched below it 
fluctuated through a wide range. In small bore the Bb saxhorn 
had a tenor quality; when the bore was enlarged and the 
taper opened up, it took on a broader quality and became 
the baritone, or euphonium; when the bore was enlarged 

and opened up still more, it was deeply sonorous and became 
the bass. 

This Bb bass, only an octave below the cornet, became 
accepted as the bass voice. Because of this, when the Eb bass, 
pitched a fifth lower, was added later, it was called the 
contrabass; likewise the Bb bass an octave below the former 
Bb bass was called the contrabass in BBb. All the upright 
saxhorns, from the Bb tenor to the BBb bass, are known as 
tubas: there are the tenor tuba, the baritone tuba and the 
various bass tubas. Today, however, it is unusual in America 
to call any instrument a tuba except the bass; in fact, the 
word tuba has become a synonym for the wind bass. This 
name tuba was taken from an old Roman instrument, but 
none of today's tubas resembles the Roman tuba, for it was a 
straight bugle only three feet long. About the only resem- 
blance between the two kinds of instruments is the wide 
angle taper, but today's tubas have even lost this. 

The baritone in Bb was sometimes called a baritone and 
sometimes a euphonium. Sometimes the difference was only 
a matter of names, but sometimes there was a difference in 
bores and tapers. Today the word baritone refers to a small- 
bore instrument while euphonium refers to a large-bore in- 
strument, both in the same Bb voice. The baritone, having 
a small bore, does not possess the big, mellow quality of tone 
possessed by the euphonium, but its tone is broader and 
fuller than that of the Bb tenor, which is still smaller in bore. 
The name baritone naturally comes from the word used to 
classify human voices, but those who christened the larger- 
bore instrument as euphonium went far afield for the name. 
The word in Greek means " sweet-sounding " and is apt 
enough from this standpoint; but it was probably stolen 


directly from another musical instrument which preceded 
it by two or three generations. In 1790 Ernst Chladni, a 
German, invented a musical instrument made of glass plates 
and rods, which also gave forth sweet sounds and was also 
called the euphonium. The wind euphonium, however, seems 
to have appropriated the name and probably has it for keeps. 
While the smaller instruments usually had only three 
valves, the euphonium and basses generally had four and 
sometimes five valves. The reason for this goes back to the 
acoustical principles of open-pipe instruments. All open-pipe 
instruments have a natural, or harmonic, scale made up of 
the following intervals: 

6 \><L -o- etc. 

In the chapter on cornets and trumpets is shown how the 
three valves serve to bridge the gaps in the scale so it is 
possible to play chromatically from the second partial up- 
ward. It is possible to descend below the second partial to F# 
by using the three valves, but this leaves a gap between F$ 
and the first partial, or fundamental. On soprano instruments 
this gap is unimportant, for the effective range of the in- 
strument is above the second partial, and while these valve 
notes below the second partial can be played, composers 
and arrangers recognize they are not very good in musical 
quality and do not call for them often. Since the valve notes 
below the second partial are not of good quality, nobody 
seems to worry about the fact that still lower valve notes are 

1*43 } 

not available on a three-valve instrument. This is true not 
only on sopranos but also on altos and tenors and baritones. 

On euphoniums and basses, however, this gap is an en- 
tirely different matter. The bore on these instruments is so 
large that the lower notes are of good quality. In fact, the 
best notes on large-bore instruments are the lower, while 
the poor notes are the top notes just the opposite of the 
soprano instruments. The valve notes below the second par- 
tial are of good quality on these larger-bore instruments, and 
it is desirable not only to play from the second partial down 
to F$ but also to play on down to the fundamental, com- 
pletely bridging with chromatic intervals the wide gap of a 
complete octave which exists between the first partial, or 
fundamental, and the second partial. But three valves with 
their six semitones are not sufficient. In descending from 
the second partial we use the second valve, first valve, first 
and second, etc., but when we have all three valves down and 
have descended chromatically as far as our three valves will 
allow us, we find outselves stranded on F#. This leaves a 
gap between the fundamental C and F# of five semitones. 
To make these notes available, a fourth valve was used 
which was capable of adding sufficient length of tubing for 
lowering the instrument five semitones. This extra valve, 
in combination with the other three valves, allows us to 
play F, E, Et>, D and Db. The next note is the open C, or 
fundamental, so chromatically we are able to bridge the 
gap of an entire octave between the first and second partials. 

This addition of the fourth valve made the euphonium 
and bass saxhorns chromatic throughout their complete 
range. Later a fifth valve was added, not because it was 
necessary to bridge the gaps in the scale but in an effort to 



correct the intonation on some of the valve notes. This dis- 
crepancy in the intonation of some of the notes played with 
the valves touches on an inherent fault of all valve instru- 
ments. A full explanation of the problem would require a 
discussion too long and involved for the scope of this book. 
The fundamentals of the problem can be illustrated roughly 
by an analogy. Suppose there were a great stairway, the 
first step of which was high and large, the next step a bit 
smaller, the third still smaller, and so on gradually, until 
at the top the steps were low and small. Suppose the lower 
steps were so high and large that a small child could not 
climb them. To assist the child, a large block half as high as 
the first step could be made which could be placed at the 
bottom of the first step. The child could step up on the 
block and from the block could reach the first step. Then 
the child could pull the block up after him and place it at 
the bottom of the second step. This block, which was made 
just half as high as the first step, would be slightly more 
than half as high as the second; and after several steps the 
block would be nearly as high as or higher than each step, 
and no block would be needed at all. 

The musical scale is like our imaginary steps: as we ascend, 
the steps grow smaller and smaller. A valve with Its tubing 
is a sort of block which helps us span a step too long for us 
to make without it. Now it is perfectly obvious that, in the 
stairway analogy mentioned above, a block large enough to 
be just half as high as the bottom step would be more than 
half as high for other steps on up, because each step is 
graduated smaller and smaller. Likewise, in music a valve 
with tubing which is just right to bridge a gap In the lower 
part of the scale would not be right for bridging a gap In the 


top of the scale. Valve tubing of proper length for the wide 
gaps in the lower part of the scale would be too long for the 
smaller gaps in the upper part; and, vice versa, valve tubing 
made the proper length for the small gaps in the upper part 
of the scale would be too short for the wider gaps in the lower 
part of the scale. For this reason, instrument makers calculate 
the length of valve slides so they will be in between and come 
nearest to a compromise up and down the scale. If the valve 
slides are all right for the middle of the range of the instru- 
ment, they throw the top a little flat because the slides are 
a bit too long; and they throw the bottom a little sharp, 
because they are a bit too short. This is more serious in the 
lower part of the scale, where the gaps are large, than it is 
in the top, where the open notes are close together. For this 
reason, some cornets and trumpets have adjustable first- and 
third-valve slides, so the slides can be lengthened when used 
for notes in the lower part of the scale. 

The problem as outlined for cornets and trumpets is 
identical with that on euphoniums and basses, only it is 
considerably augmented in the case of the latter. In our 
imaginary stairway, it is easy to see that if we should add 
one or two steps at the bottom, each correspondingly larger 
than those above it, the discrepancy between each step would 
rapidly increase. Therefore the inadequacy of valve slides 
on large instruments like the euphonium and bass is greatly 
magnified. Valve slides which were of proper length in the 
middle of the scale would be far too long for the top and still 
farther too short for the bottom. For this reason, the fourth 
valve on basses and euphoniums is especially welcome, not 
only to bridge the gap between F# and C but also to use as 
an alternative for the combination of the first and third 



valves. The fourth valve is roughly equal to five semitones, 
but it usually is made somewhat longer than the combined 
length of the first valve (with its two semitones) and the 
third valve (with its three semitones). 

Although the fifth valve is rare on American basses, 
double-bell euphoniums usually have a fifth valve, not for 
the purpose of affecting intonation but to control the smaller 
bell. The instrument normally uses the larger bell, but a 
fifth valve can be operated to bring the smaller bell into use. 
The smaller bell is used for trombone and echo effects. 

Although our family of tubas is organized on the same 
basis as the saxhorns, it did not descend directly from the 
saxhorns but from the saxtrombas. This family of instru- 
ments, which is also an invention of Sax, was brought out 
about 1850, eight years after the saxhorns. There were eight 
members in this family, including the Eb sopranino, the Bb 
soprano, the Bb mezzo-soprano, the Eb alto, the Bb tenor, 
the Bb baritone, the Eb bass and the BBb contrabass. The 
saxhorns were conical in bore and were played with a deep 
bell-shaped cup mouthpiece. These features produced a 
broad, mellow tone. The saxtrombas were designed with 
smaller tubing, about one third of which was straight, or 
cylindrical, the other two thirds being conical. The mouth- 
piece was not so deep as that of the saxhorns and was shaped 
more like a bowl. The resulting tone was brighter and more 

Most European countries prefer the saxhorn family be- 
cause of its mellow tone. In Italy and England especially, 
the conical-bore instruments are used almost universally. 
In America the saxtromba family is preferred, or at least 
the modified saxtrombas which we have developed. The 


American instruments are larger in bore than the original 
saxtrombas and are played with a more shallow mouthpiece. 
However, the fundamental principle in the bore construction 
has been followed, the bore having less taper than the bore 
of the saxhorn. While these later instruments of Sax's had 
no wide acceptance in Europe, they have come to be, in 
their slightly modified form, America's choice. 

There were nine distinct members of this family which 
grew up on American soil The small Eb cornet used to be 
quite a popular instrument but has fallen into disfavor 
lately. The Bb cornet is the favorite soprano of this big 
family. These two instruments are played in horizontal 
position, but the rest of the family are upright instruments. 
They include the Eb alto and the Bb tenor, both of which 
are now almost obsolete; the Bb baritone; the Bb euphonium; 
the Bb bass, which is entirely obsolete; the Eb bass and the 
BBb bass. 

Along about the time of the Civil War a peculiar family 
of tubas was quite popular. This was the instrument which 
"backfired " over the left shoulder. The family was made up 
of soprano, alto, tenor, baritone and bass. All of them were 
played with rotary valves, rare in America on any instrument 
except the French horn. The idea behind their peculiar 
appearance was that the bell pointing backward over the 
shoulder would throw the music of the band backward. 
Since the band marched at the head of the army or parade, 
this seemed a logical idea. After the war, the military spirit 
died out, marching bands gave way to concert bands, and 
the backfiring instruments gave way to those with upright 
bells or bells projecting forward. They are now only interest- 
ing relics. 



Although not belonging to the family of tubas, but rather 
to the French horns, the Wagnerian tuben were an interest- 
ing, if short-lived, invention. Wagner introduced these in- 
struments in his "Nibelungen Ring" as a means of extending 
downward the French horn range. He apparently was not 
satisfied with the tone quality of the regular tubas as basses 
for the French horns, and he had the tuben built as an ex- 
periment. There were two a tenor in Bb and a bass in F 
and they were played with a deep funnel-shaped mouthpiece 
like a very large French horn mouthpiece. There is no evi- 
dence that Wagner was not satisfied with his invention, but 
these instruments were not widely used by anyone else and 
soon became obsolete. Occasionally tuben are used in these 
Wagnerian numbers, but usually the tuben parts are played 
by the regular tubas. 

Until after the Civil War there were no important tuba 
makers in America. The instruments were imported from 
England, Germany, France and Italy. Henry Diston, a 
famous cornet soloist from England, toured America with 
his three brothers shortly after the Civil War and finally was 
persuaded to head a band instrument firm in this country. 
The firm was established in Philadelphia, and although the 
instruments were supposed to be made in America, many 
were imported and merely stamped with the Diston name. 
Pepper also started a band-instrument factory in Philadel- 
phia shortly after the war. Hall Bros, of Boston, inventors of 
the "pinched rotary valve," and Isaac Fishe of Worcester 
were other makers. The bulk of the instruments used in 
American bands, however, continued to be imported. 

The split lip of a cornet player in 1873 started the band- 
instrument business in America. Captain C. G. Conn re- 

turned to Elkhart, Indiana, after the Civil War and started 
a grocery and bakery. He was a musician and played cornet 
in the silver-cornet band. One day the Haverly Minstrels 
visited the town, and during an altercation with a big bass 
player in the visiting troupe Conn received a badly split lip. 
This prevented him from playing the cornet, and to help 
remedy the situation he invented a mouthpiece with a rubber 
cushion. The invention became known, and before long Conn 
found himself in the mouthpiece business, making his mouth- 
pieces on an improvised lathe made from a sewing machine. 

In 1875 Conn started making cornets, with the assistance 
of a partner by the name of Dupont, an English horn maker. 
Experienced horn makers were secured from England, France 
and Germany. Early in his experience Conn obtained patents 
on several inventions, among them being front-action instru- 
ments with valve slides ascending from the valves and pre- 
venting water from entering the slides. This new family of 
tubas was called the "American Band Instruments," and it 
wasn't long before they were competing on a favorable basis 
with imported instruments. The first double-bell euphonium, 
now so familiar, was built by Conn in 1890. In 1898 Conn 
built the first sousaphone, a variety of helicon with the bell 
opening upward. The instrument was built especially for 
Sousa and was named in his honor. In 1908 the first bell-front 
sousaphones were made and thereby hangs many a tale of 
mirth for the initiated. 

The ignorance of the history of the sousaphone seems to 
be universal among movie technicians. It is a favorite sport 
of the few who know when this instrument was first created 
to watch the movie productions for uses of these modern 
instruments in historical scenes. British soldiers bound for 


(Left to right, cross sections) : Mouthpiece of saxhorn and mouth- 
piece of modern tuba. Note that the saxhorn mouthpiece is deep 
and funnel-shaped* while the tuba mouthpiece is more bowl- 
shaped. This difference in shape in the mouthpiece has much to 
do with the difference between the tone quality of these two types 
of instruments. The saxhorn is more mellow while the tuba is more 
brilliant in tone quality. 

Carried by Timothy Church of Connecticut in the famous battle^ 



the Boer War are serenaded by a band with a battery of 
bell-front sousaphones, at a time when there were only four 
sousaphones in existence and these had upright bells and 
were all four in Sousa's band. Spanish-American War fervor 
is made to live again in music from the sousaphone, at a time 
when no sousaphone had yet been built. The Boys in Blue 
and the Boys in Gray are both cheered on to victory or are 
honored in celebrations of triumph by bands whose bass parts 
are played on sousaphones, at a time when Captain Conn 
was merely a soldier and over a decade before he ever 
thought of making a band instrument. The gold rush to 
California, the completion of a railroad track by driving a 
gold spike, packet-boat races on the Mississippi are all occa- 
sions for parading sousaphones before the public, and nobody 
seems to worry about the fact that the first sousaphone was 
not made until 1898 and the bell-front sousaphone was not 
made until 1908. 

Although the bell-front sousaphone is the most popular 
bass tuba in American bands, Sousa steadfastly clung to the 
original sousaphone with bell up whether for musical 
reasons or for sentimental reasons, no one knows. Either 
would be sufficient justification in this instance. When Sousa 
died in 1932 and his band was dissolved, the four sousaphones 
were bought as noble souvenirs of a noble man. One of these 
original sousaphones from Sousa's band is now exhibited by 
the Museum of Science and Industry in Chicago. And as 
spectators gaze upon this interesting specimen they can 
reflect that they are gazing upon the latest offspring of the 
prolific serpent although, if the strain runs true to form, 
certainly not the last. 


Percussion Instruments 


WHEN THE MOORS CROSSED from, Africa into Spain in 
711 A.D. they brought with them some queer drums 
which looked like the two halves of a large ball. Five hundred 
years later the Crusaders, returning from the Orient, rode 
across Europe with these same half-sphere drums slung across 
the necks of their horses. Today we have the giant kettle- 
drums, or tympani, which trace their ancestry directly to 
these ancient instruments. 

It cannot be said that either the Moors or the Crusaders 
introduced drums into Europe, for the drum is an instrument 
which is indigenous to every soil and climate. There are no 
peoples on the globe, however primitive or cultured they 
may be, who do not play the drum in some form or other. 
There has never been a time in the history of the world in 


which the drum did not figure prominently.fToday the wild- 
est tribes in the heart of Africa beat their drums to accom- 
pany the ceremonies of war and death, joy and celebration. 
Today in the finest concert bands and symphony orchestras 
all over the world the drums boom out to sound a rhythm 
or punctuate a musical phrase./ 



The warp and woof of music are rhythm and melody, and 
the drums are the rhythm instruments par excellence. It is 
easier to recognize a song by its rhythm without melody than 
it is by its melody without rhythm, which shows what a 
basic part of music is rhythm. Primitive music is more 
rhythm than it is melody, Some of this primitive music is 
tremendously expressive. Melody could add very little to the 
foreboding pulsations of the African war drums. In fact, 
melody would detract more than it would add. There is 
something in the constantly recurring rhythmical beat of the 
drums which pulsates in the blood. There is something in the 
incessant and ominous boom of the drums which pounds in 
the brain. Melody would relieve the tension, would break the 
spell. But the dread rhythm of the war drums, beating in the 
ears, booming in the brain, speaks a terrible message which 
could be spoken in no other way. 

If it be a dirge, how little is melody missed when the drums 
begin their lament! With a rhythm peculiarly expressive of 
grief and sorrow, the drums beat out a mournful elegy which 
asks nothing of either words or melody. By contrast, what 
can be gayer than the castanets and tambourines of Spain 
or the bongas and maracas of Cuba? The quickened rhythm, 
the joyous accents of these instruments sing a song of gaiety 
and happiness which melody could scarcely supplement. 
What can the melody of the bugle add to the stirring rattle 
of the military drum, sounding assembly or commanding a 
charge? The weird, the mysterious, the terrible all can be 
portrayed with tremendous drama and reality by bare 
rhythm without melody. 

It is no wonder that all peoples, from the most primitive 
and barbarous to the most educated and cultured, have been 

I 2 S3l 

lovers of the drum and other percussion instruments.* In 
earliest history we learn that the Egyptians, Assyrians, 
Hebrews, Greeks and Romans all used instruments corre- 
sponding to our kettledrums, tenor drums, tambourines and 
cymbals. Of these, the most important soon came to be the 
kettledrums. In early Europe they were used not only in 
military affairs, but in the court of Edward I as musical 
instruments. Later, in 1347, when Edward III celebrated his 
triumphal march into Calais, kettledrums helped make the 
music. Chaucer often speaks of the "nakers" in his Canter- 
bury Tales, and nakers is an Arabic word meaning "kettle- 
drums." In a carving in Worcester Cathedral, believed to 
have been done in 1396, a pair of kettledrums is shown 
strapped to the waist of a player, one on each side. These 
were small kettledrums, similar to those brought by the 
Moors into Spain and carried by the Crusaders from Arabia, 
but larger-size kettles were developed by the Germans, 
which are practically like our modern tympani. Henry VIII 
introduced these larger kettledrums into England in the 
first half of the sixteenth century. The German historian of 
music, Virdung, writing in 1511, describes the kettledrums 
of his day. He even draws some pictures of them which look 
much like the modern kettledrums. About a hundred years 
later, Praetorius, another German historian of music, talks 
about the kettledrums; and so does the Frenchman Mer- 
sennus, writing in 1627. These ancient kettledrums were 
hemispherical and had skin heads stretched across the top 
by hoops which were held in place and tightened by adjusting 
screws around the rim. ,3r 

Kettledrums graduated from the army and the military 
band into the orchestra during the time of Lully and were 



used commonly by him and other French composers of the 
seventeenth century. As early as 1713 kettledrums had be- 
come popular in Germany, for Johann Mattheson, of Ham- 
burg, composer and musical authority, writing of the musical 
instruments of his day, says that kettledrums were often 
used in both church and opera. These he says were used in 
pairs and were tuned a fourth apart, a practice which existed 
for many years. Handel knew about kettledrums, using them 
in his "Water Music." Bach also used them, as did Haydn 
and Mozart and all the other great masters who came 

These early kettledrums, or tympani, as they are now 
called, were hand tuned and were pitched in C and G, the 
tonic and dominant of the key in which the music was 
written. The large kettle was tuned to the G below the C, 
while the small kettle was tuned to the C, making them a 
fourth apart. The reason for this inversion was the limita- 
tions of the instruments. If the tonic had been given to the 
large kettle and the dominant to the small kettle, the 
dominant would generally have been higher than the small 
kettle's compass. Therefore, the tonic was given to the 
small kettle, and the dominant an octave below was given 
to the large kettle. Kettledrums were treated mostly as 
military instruments, for they were hardly ever allowed to 
play except with the trumpets, in marches, overtures and 
other such music. This is only another example of following 
custom. Trumpeters and kettledrummers used to accompany 
royalty wherever it went and were used to signify rank, 
much as rank is signified today by cannons, a certain number 
for each rank. Later, when trumpets were admitted to the 
orchestra, the kettledrums naturally followed; also, when 


the trumpets played, the early composers thought it appro- 
priate that the kettledrums play, too. 

It was Beethoven who freed the tympani from these 
shackles, not only those imposed by the custom of pairing 
the kettledrums with the trumpets, but also the universal 
tuning to G and C, a fourth apart. In his First Symphony in 
1800, Beethoven startled the tympani player and the audi- 
ence by having the tympani play a sort of bass part to a 
melody of violins and flutes. Seven years later, in his Fourth 
Symphony, he elects the tympani to the great honor of 
stating a theme of two notes which was repeated by the 
other instruments. The following year, in his great Fifth 
Symphony, the same symphony in which the piccolo, trom- 
bone and contrabassoon all make their debut in the sym- 
phony, Beethoven causes the tympani to make their debut 
as a solo instrument, creating for the tympani a solo effect 
in the scherzo movement. In 1814, in his Eighth Symphony, 
he tries still another innovation by having the tympani 
play in unison with the bassoons. By this time the fatal tie 
between the Siamese twins had been broken and the tympani 
was no longer restricted to duets with the trumpet. 

In tuning, also, Beethoven experimented with the tympani. 
He first dared to tune the kettles in fifths, giving the large 
kettle the tonic C and the small kettle the dominant G 
above. This is common today, but in Beethoven's time it 
was a daring change. Encouraged by this success, in his 
Seventh Symphony, composed in 1813, he tuned the two 
kettles in sixths. In his Ninth Symphony, composed eleven 
years later, he positively went berserk by tuning the tympani 
in octave F's and introducing the unheard-of practice of 
playing two notes of a chord simultaneously on two kettles! 


Not only did he make innovations in tuning and in ways 
of using the tympani, but he demonstrated their resources in 
musical effects. He showed that the tympani could be used 
not only for brilliant and military effects, but also to express 
mystery, solemnity and awe. He established them as versatile 
and useful instruments for many roles. 

The composer Meyerbeer was at one time tympanist under 
Beethoven, and he appreciated more fully than anyone else 
how Beethoven had liberated the tympani, and when he 
began composing he carried with him the open mind of 
Beethoven. He excelled his master in this respect by writing 
for four tympani instead of the conventional two, tuned in 
G, C, D and E. This novel bit of tympani writing occurs in 
Meyerbeer's "Robert le Diable," written in 1831. 

A countryman of Meyerbeer, Hector Berlioz, went even 
further; in fact he went to extremes. In his "Requiem Mass" 
he startled the musical world by writing for eight pairs of 
tympani tuned as follows: minor third, minor sixth, major 
third, fourth and diminished fifth. These sixteen tympani 
were played by ten players, some of them playing on pairs 
and others booming away on a single tympano. In this 
effort he was attempting to portray the crack of doom on 
the great resurrection day, and the effect was appalling. 

In reviewing the practice of his day, Berlioz writes, in his 
great book on instrumentation and orchestration, that it was 
customary to tune the tympani in a great variety of ways : 
minor third, major third, second, perfect fourth, augmented 
fourth, fifth, sixth, seventh and octave. He also recites how 
he introduced the practice of having one player use three 
tympani instead of two. He says this occurred while he was 
directing an orchestra in the Paris opera house. He points 

out that it required seventy years of tympani playing in the 
orchestra to develop the simple idea that one tympanist 
could easily play three tympani instead of two. In this he 
was in error, however, for Weber in 1807 had used three 
tympani in his overture to " Peter Schmoll," and three tym- 
pani had also been used by Mendelssohn and by Schumann. 

Wagner was partial to the tympano and raised tympano 
writing to new heights. Not only did he write some beautiful 
solos for the tympani, as in the introduction to "Siegfried," 
but he made of the tympano a great dramatic instrument. 
He prolonged the suspense before some tragic happening, 
by using the soft, foreboding strokes of the tympani. Or, 
when the whole orchestra was hushed in some tense moment, 
he called for a soft note on the tympani to magnify the 
stillness. Classical examples of this dramatic casting of the 
tympani are in " Lohengrin/' where Talramund drops dead 
at the sight of the holy word of the Grail Knight whom he 
had wished to kill, and also in "Gotterdammerung" after 
Siegfried is stabbed by Hagen. 

With this greater taste in writing for the tympani, the job 
of the tympanist became less difficult in one way and more 
difficult in another. He used to have to play whenever the 
trumpets played. He might be called on to boom his way 
through an overture or other brilliant music from the opening 
beat until the final crash. But after the new style of writing 
he might be called on to play a couple of short strokes, and 
then he'd have to wait through many measures for another 
couple of notes. In Rossini's "Semiramide" the tympani 
have a tremolo in the first measure, and then follow seventy- 
eight measures of rest. Schubert's Seventh Symphony is 
noted also for its many measures of rest. Some tympanists 



depend on a cue from the conductor, but many count the 
measures and can come in accurately on an afterbeat follow- 
ing dozens of measures of rest. In fact, among tympanists 
this is a point on which to boast. Some claim they can go 
out for a smoke and come in just in time to pick up the 
mallets and play their note or roll. 

An amusing story on this point is told about Pfund, the 
famous tympanist under Mendelssohn. He was a wizard on 
counting the silent measures and never failed to come in on 
the right beat. Once he loaned a bass viol player a thaler. 
Soon after, he began dunning the viol player to pay it back. 
The viol player was so pestered by Pfund that he decided he 
would pay it back and at the same time get revenge on 
Pfund for his illiberal attitude. One day when Pfund was out 
for a smoke during many measures of rest for the tympani, 
the viol player laid pfennigs equal to one thaler around the 
edge of one of the tympani. As usual Pfund came in just in 
time to take up the mallets and strike a resounding boom 
on the instrument. This blow on the tympano caused the 
pfennigs to jump into the air and come down in the head 
again, dancing, jingling and rolling to the floor, much to 
the consternation of Pfund, Mendelssohn, the rest of the 
orchestra and the audience. 

This prank may or may not have been the beginning of 
novel effects on the tympani made by laying keys or metal 
plates on the head, but these effects are now written legiti- 
mately for the tympani. Certain it is that Pfund taught 
future generations of tympano players to play with artistry. 
He is regarded as the first great artist on these instruments. 
Mendelssohn took advantage of his artistry to write many 
difficult passages for the instruments, and succeeding tym- 


panists had to be artists to play the music written especially 
for Pfund. Mendelssohn called for many tunings of the 
tympani which were queer at the time, among them C# and 
A, D and E, G and F, Bb and Db. 

But the manufacturers of tympani have made it possible 
for modern composers to write music for the tympani which 
even Pfund could not play. Pfund's tympani were hand 
tuned, and to change the tuning of the instruments required 
a certain interval of time and a most accurately trained ear, 
especially when the tuning had to be changed during a num- 
ber. Pfund apparently was distressed with this shortcoming, 
for he made the first attempt to construct tympani which 
could be tuned by a foot pedal. The first set he built is still 
in the museum of the Civic Opera House in Leipzig, Ger- 
many. They are crude, iron forgings and were clumsy to 
operate. Now, however, through the use of the pedal or 
machine tympani, the tuning can be quickly changed by a 
lever worked with the foot. Merely by working the pedal 
up or down, the pitch of the separate kettles can be changed 
to any pitch within the range of a fifth or sixth. The large 
kettle usually has a range of from F to C, and the small 
kettle has a range of from B to E or F, giving an overall 
range of a full octave. So quickly can the tuning be changed 
on these modern tympani that it is possible to play a tune 
of somewhat complicated melody on the tympani! Even 
Pfund would have been amazed at this. 

The tendency in writing for the tympani is to make them 
more and more solo instruments. Some of this advanced 
writing is found in the works of Ravel, Strauss and Stravin- 
sky. Ravel, in his famous "Bolero," gives to the tympani 
the complete bass accompaniment toward the end. 



Requisites of a good tympanist are a most refined sense of 
rhythm and timing and a keen ear for tuning. The excellent 
qualities of Pfund in accurate timing are still as needful 
today as ever, and although tuning has been greatly facili- 
tated through the pedal tympani, no less keenness of pitch 
is required. There are no stops for the pedal to indicate 
different notes, as there are frets on the guitar, and the 
tympanist must still rely on his ear to tell when the right 
notes are obtained. The pedal merely stretches the head 
tighter to obtain higher notes and loosens it for lower notes, 
but the tympanist's ear must tell him when the right notes 
are obtained. 

In the old days the tympanist would throw a cloth over 
the heads of the tympani to muffle the sound for certain 
soft effects. Today the manufacturers have supplied him 
with an assortment of mallets with hard or very soft heads 
which enable him to play with brilliance or with muffled tone, 
as the music requires. Different qualities of tone can also be 
obtained, depending upon where on the head, with reference 
to the center, it is struck. The head is usually struck midway 
between the outer edge and the center. Striking the head near 
the center produces a tubby tone, while striking it near the 
edge produces a bright but thin tone. 

It is the belief of some that the pitch of the kettle is deter- 
mined by its shape and size. This has been disproved by 
filling the kettle with water. It is true that the diameter of 
the kettle affects the pitch. The lowest note on the standard 
twenty-eight-inch kettle is F or possibly E, and for some 
classical compositions requiring lower notes, a special thirty- 
inch kettle must be used. Gustav Mahler wrote low D in 
his Ninth Symphony, for which a special thirty-five-inch 

kettle must be used. Also, the highest note for the standard 
twenty-five-inch kettle is F or possibly G, and to play notes 
higher than this, a special twenty-four-inch kettle must be 
used. Rimsky-Korsakov wrote high Eb in his "Mlada," for 
which a special twenty-two-inch kettle must be used. 

Percussion instruments are conveniently classified as to 
those which have definite pitch and those which do not. 
The tympani have definite pitch, as has been related. Other 
such instruments are the family of bells and chimes. The 
bell family comprises many instruments, which are played 
in a variety of ways. Some are made of bars of metal, such 
as the orchestra bells, glockenspiel and celesta. Some are 
made of bars of wood, such as the xylophone and marimba. 
Some are struck in the center of the bar with mallets, as the 
orchestra bells, glockenspiel, xylophone, marimba. Some are 
made of tubes of metal which are struck on the end, as the 
chimes. The celesta is played by pressing down a key 
mechanism, the same as the piano. 

If it weren't for the fact that bars of wood or metal were 
used in musical instruments a long time before the Crusades, 
the interesting story about Saladin might be used to explain 
the creation of the resonating bar. The story is a romantic 
one, however, and it is really too bad that it is in disagree- 
ment with the facts. When Saladin, Emperor of Turkey, 
conquered Jerusalem in the latter part of the twelfth century, 
he commanded that all bells be broken up and destroyed 
so the Christians could not be called together for worship. 
Bars of wood and iron were made and substituted for the 
purpose. These took up smaller space and could be more 
easily secreted from the eyes of the infidel. 

Economy of space and greater convenience are responsible 



for the substitution of the modern tubular chimes for genuine 
bells. In the Middle Ages, carillons of bells were one of the 
important sources of music. At first these carillons were 
comprised of three to six bells which were struck by hammers 
held in the hand. Later a dozen or more bells were used, and 
they were struck by a system of ropes pulled from below the 
cathedral tower. Today these giant carillons are played from 
a keyboard, like an organ, the bells being struck by a mecha- 
nism actuated by the keys. Two of the largest carillons in the 
world are those at the University of Chicago Chapel and at 
Riverside Church, New York. The largest bell weighs over 
eighteen and a half tons, and the smallest weighs ten pounds. 
In 1925 the largest bell was found in the fifty-two-bell set at 
Ghent. This bell is ninety-eight inches in diameter and 
weighs 20,720 pounds. Its pitch is E below Middle C. 

Composers have often wished to produce the sound of 
these carillon bells, but it is obvious that they could not 
install one or several of these actual bells, because of their 
tremendous size and weight. It was found that metal tubes 
one inch to two and a half inches in diameter and only four 
to six feet long would reproduce the pitch and tonal quality 
of all but the lowest notes of these giant carillons. It is chimes 
of this type which are used in the modern orchestra to create 
such bell effects as are heard in Tschaikowsky's "1812 Over- 
ture" and in Wagner's "Parsifal." The standard chimes are 
built with eighteen tubes and have a range from an octave 
above Middle C chromatically up to F, an octave and a 
fourth above. 

Orchestra bells are a series of metal bars of graduated 
length and thickness, which give a chromatic range of two 
octaves or more. They are made with or without resonators, 


are struck with hard mallets, and give off a bright, metallic 
sound. They are heard in Wagner's "Die Walkiire," in the 
magic fire scene, and also in his "Die Meistersinger," in the 
last act. 

The xylophone is a similar instrument except it has a 
range of three and a half to four octaves and the bars are 
made of rosewood instead of metal. This wood is very hard, 
and when struck with hard mallets it produces a brilliant 
sound of short duration. Some composers regard the xylo- 
phone as crude, but Saint-Saens has used it to advantage 
in his famous "Danse Macabre," where he depicts the 
rattling of bones in the death scene. 

The marimba is similar to the xylophone, but the brilliant 
sound of the xylophone has been deepened and prolonged 
by the use of resonators which build up and enrich the tone. 
These tubes are placed beneath the rosewood bars and are 
of varying length in proportion to the varying length of the 

A more modern creation is the vibraphone, a sort of ma- 
rimba with metal bars and an electrically actuated valve 
which rotates in the top of the resonator tube, producing a 
pleasing vibrato. 

To Tschaikowsky goes the honor of having introduced 
into the symphony the celesta, a type of orchestra bells 
played with a keyboard. This interesting bit of music is 
found in his "Casse-Noisette." 

In the band a transportable form of orchestra bells is used, 
called the glockenspiel, or bell lyra. The metal bars are made 
of special aluminum alloy and are mounted on a frame 
shaped like a lyre, while the standard for the lyre is carried 
in a belt like a flag or banner. In marching bands the bright, 



piercing notes of the bell lyra can be heard above the band. 
Recently these instruments have become popular in the 
band at football games, and when the band music is broad- 
cast, these instruments can be distinctly heard when the 
rest of the band fades to a whisper. 

Everyone is familiar with the bass drum, the side drum or 
snare drum, the cymbals and the triangle, all instruments of 
indefinite pitch. Although they were used in military bands 
for a long time previously, Gluck is credited with introducing 
them into the orchestra about 1760. Desiring to create some 
special effects in his operas, he called in these lowly musical 
instruments as an experiment, and they have remained ever 
since. Mozart used the bass drum, cymbal and triangle a 
few years later, to create a Turkish atmosphere in his "Die 
Entfiihrung." During the last quarter of the eighteenth 
century these instruments were used often when the com- 
poser wished to portray the oriental or barbaric, and Haydn 
used them once, in his "Military Symphony.' 7 Before the 
middle of the next century, most large orchestras in opera, 
such as those in Berlin and Paris, counted the bass drum, 
cymbal *and triangle as standard equipment. They were also 
usually found in ballet and light opera, but were not ad- 
mitted to the more exclusive ranks of the symphony until 
some time later. 

The old side drum became the snare drum after snares 
were added. In England this drum is still usually called the 
side drum, but in America it is more often called the snare 
drum. This name comes from the "snares" which are 
stretched across one head, inside the drum. The creation of 
this device is credited to a Scottish drummer. This drummer 
originated the use of a rawhide whip for striking one side 


of his bass drum, and sometimes he would hold the whip 
against the head on one side while he struck the drum with 
the drumstick on the other side. This created a sort of dry 
rattle. Later, catgut strings were stretched across the head, 
and these were called snares. It is this device which creates 
the snappy, rapid vibration of the snare drum when it is 

The tambourine is a very ancient instrument, being pic- 
tured in the ruins of the Egyptians, Assyrians and Greeks. 
It is now closely associated with the Spanish, due to their 
love for and wide use of this instrument. One of the first 
uses of the tambourine in the orchestra was by Weber in his 
overture to "Preciosa." 

The name castanets seems to have been derived from 
"castana," the Spanish word for chestnut, which would 
indicate that originally they were made of this wood. Some 
authorities claim the Moors introduced castanets into Spain, 
but their instruments were made of metal. Now castanets are 
made most often of vulcanized rubber, called ebonite. This 
is also a Spanish instrument, and it is claimed that few 
people not Spanish-born ever learn to use it properly. The 
most celebrated use of the castanet is in Bizet's opera, 

Then there are tam-tams and tom-toms. These two in- 
struments are often confused by the uninitiated, and it is 
to avoid this that the tam-tam is now more generally called 
the gong. It is a huge cymbal suspended in a metal ring so 
the instrument can vibrate freely. It is struck with a mallet, 
and it resounds with a prolonged reverberation. The tom- 
toms are merely modern revisions of the ancient and primi- 
tive tom-toms of the savages. They are small drums with 



thick leather heads, and they produce a sharp, brief, hollow 
sound similar to that of the more familiar bass drum, only 
more shallow and of less musical quality. They are seldom 
used in the orchestra but are popular in the dance band. 

The bass drum, snare drum, cymbal, triangle, tambourine, 
castanet and gong are of indefinite pitch. Although they 
cannot be used for melody, they are essential in many ways 
to set tempo, emphasize musical dynamics and give color. 
Although not a pitched instrument, the bass drum is tight- 
ened so it tunes with low F for best quality of tone. There is 
no definite pitch to cymbals, but they should have a distinct 
quality of tone. The best cymbals are made in Turkey and 
are hand hammered from bronze by a process which no 
other makers have been able to duplicate. Chinese cymbals 
are hand hammered, and American cymbals are spun, but 
neither of them has the rich quality of the Turkish cymbal, 
although American manufacturers have made important 
progress in recent years, particularly in matched pairs. 

Considerable variation in tone color can be made by the 
player in the way this group of instruments is played. The 
bass and snare drummers do more than beat or strike their 
drums. They have an elaborate and intricate technique 
which is appreciated only by those familiar with drumming. 
Cymbals, too, must be played properly to obtain the utmost 
from them. One of the most dramatic actions in the sym- 
phony or band is the glancing stroke of the cymbals by the 
good cymbal player. This looks easy, but great skill is re- 
quired to make the cymbals kiss each other at the right place 
and with the proper touch as one hand goes up and the other 
goes down. 

Haydn set a precedent in his "Seasons'* which he probably 


would regret if he were alive. He introduced "effects" such 
as thunder, the whistle of the quail, a gunshot, the chirp of 
the cricket, and today there is no end to the "effects" which 
are used in the symphony orchestra, not to mention the 
effects used in the theater orchestra! To leaf through a 
modern drum catalog is something like going to a carnival 
or visiting a freak show. No one becomes particularly upset 
by such inoffensive traps as temple blocks, wood blocks, 
slapsticks, and jingle clogs. An eyebrow might be raised at 
the cyclone whistle, locomotive imitation, steamboat whistle, 
boatswain whistle, fire bell or police gong. One might wonder 
a little at such old livery-stable effects as sleigh bells, horse 
hoofs and horse neighs. Some misgivings might be expressed 
when considering such Cuban gadgets as claves, maracas, 
bongos, and perhaps Haydn would have been amused at the 
cuckoo call, the bird whistle, the rooster crow, the dog bark, 
the crow call, and even the cow bawl. But we reach the ut- 
most in something or other when we come to such sound 
effects as hen cackle, fly buzz, pig grunt, baby cry, duck 
quack, monkey chatter, snore imitation and nose blow! 

The number of men in the percussion section varies con- 
siderably according to the composition being played. Large 
orchestras use from four to eight players. The concert band 
as a rule consists of three and sometimes four men, but occa- 
sionally a small drum corps is added for playing certain 
military music and the number is greatly augmented. The 
average theater orchestra employs two percussionists, one to 
play "trap drums," a foot-pedal device for beating the bass 
drum while at the same time he uses both hands to play the 
snare or side drum. This player is sometimes referred to as 
the double drummer. The other player, assigned to the 



tympani, ulso doubles on bells, marimba or xylophone. Only 
one playe: is used in the percussion section of the dance band, 
but he must be extremely versatile, for he plays practically 
everything in the drum catalog snare drum, bass drum, 
bells, marimba, xylophone, chimes, tom-toms, wood blocks, 
cymbals, gong and any of a hundred other "effects." 

The percussion section has long been the stepchild of the 
orchestra and especially of the band. It was looked upon as 
something which was customary to have, but was seldom 
understood by the composer and still less by the conductor. 
The drummers were turned loose on their own, and many 
conductors and bandmasters knew so little about the art of 
the percussionist that they never knew whether the player 
was reading the music properly or not. Now, however, great 
interest is being taken in this neglected section, and musical 
performance is greatly improving as a consequence. Drum- 
mers must play the "rudiments" strictly and not "fake." 
Moreover, new effects are being tried, and the value of such 
instruments as marimba, xylophone and vibraphone are 
becoming recognized. In many respects, the percussion sec- 
tion is a virgin field of music, and great strides in music are 
being made here and will continue to be made. 

The greatest bandmasters are those who best understand 
the percussion section. John Philip Sousa knew drumming 
thoroughly, and much of the secret of his stirring marches is 
effective drum parts. Nothing in Sousa's drum scores is left 
to guess: every beat, tap and roll is meticulously written out, 
and woe to the Sousa drummer who missed a single detail. 
Sousa's trained ear was sure to catch the slightest deviation 
from the carefully worked out score. Sousa greatly valued 
his snare drummer Tom Mills and often said Mills was the 

best percussionist he ever heard. Edwin Franko ^Goldman, 
director of the famous Goldman Band, insists that v the bass 
drummer is the most valuable player in his band Frank 
Simon, director of the popular Armco Band, regards George 
Carey, the snare drummer and tympanist, as the life and 
snap of the band. i 



Plucked and Struck Strings 


y4T FIRST THOUGHT, two less-related creations can scarcely 
/JL be thought of than the fine concert piano and the 
suspension bridge. And yet their developments are but two 
chapters in the same fascinating story of steel wire drawn 
through a die. Without drawn wire we could have neither the 
piano nor the suspension bridge, for wire is essential to both. 
The ancients had their psaltery, lyre, lute and harp, strung 
with "catgut 5 * or vegetable fiber, as well as their suspended 
footbridge, but the concert piano and the Brooklyn Bridge 
are so far superior as to be almost different in kind. 

Wire is such a common article today that it is difficult to 
realize that we have not always had it. But there is no record 
of drawn wire before the eighth or ninth century A.D. It is 
true that flat or cut wire is as old as history. One of the oldest 
references to this kind of wire is found in Exodus, Chap. 39, 
verse 3 : "They did beat the gold into thin plates and cut it 
into wires." The Assyrians and Babylonians in 1700 B.C. 
and the inhabitants of Nineveh in 800 B.C. made wire by 
the same process. Most of such wire was made of gold and 

silver and was used as ornament. But the first record we have 
of round wire drawn through a drawplate is found in the 
writings of an ancient monk by the name of Theophilus, a 
German or Anglo-Saxon of the eighth or ninth century. 

Theophilus' description of the process shows that it was 
essentially the same as that used today. He mentions two 
pieces of iron "three or four fingers wide" and having three 
or four rows of holes in them. Small bars of metal were drawn 
through these holes, elongating the bar into round wire. As 
the wire was pulled through the smaller holes, the diameter 
of the wire became smaller and the length became greater. 
Such wire was soft and not very strong, because it is obvious 
that the wire must be softer than the metal drawplates, and 
at that time the hard steel of today was unknown. One 
formula mentioned by Theophilus was two parts of tin and 
one part of lead. Such wire lacked the tensile strength suit- 
able for the modern piano and even for the old clavichord 
and harpsichord. Some of it was used for making coats of 
mail for the knights of old, and no doubt many of the Cru- 
saders rode to Palestine in coats of mail made from such wire 
as is described by Theophilus. 

In 1270 we find in Paris a small guild of wire drawers. We 
do not know how long this little industry had been in exist- 
ence, but we do know that it did not thrive or grow very 
rapidly, for there were only eight such craftsmen in Paris in 
1292, and in 1500 there were only nine. No doubt one reason 
for the small number of master wire drawers was the almost 
prohibitory system of apprenticeship, for an apprentice was 
required to work ten years for nothing and had to pay a fee 
of twenty sous besides. The work of these apprentices was 
also strenuous, for they had to pull the wire through the plate 


by muscular strength or by the use of a spool which was 
turned by hand. 

The date for the beginning of wire drawing which is usually 
given is 1350, and this is the date given in Grove's dictionary. 
This reference is to the wire drawers of Augsburg, Bavaria. 
There was also wire drawing at this time in Nuremberg, 
another city of Bavaria, and in Altena, in the district of 
Westphalia. But these German wire drawers had been pre- 
ceded by about five hundred years by the craftsmen men- 
tioned by Theophilus. It is true, however, that Germany 
became the principal source of wire for the builders of clavi- 
chords and harpsichords all over Europe. Industrial England 
did not start drawing wire until 1565 and up to as late as 
1800 imported from Germany most of its piano wire. In 
America the wire industry was started in Lynn, Mass., in 
1665, but most of this wire was made for carding textiles. 
It was not until 1850 that the first piano wire was made. At 
that time Jonas Chickering, the piano pioneer, asked the 
firm of Washburn & Godard of Worcester to make some piano 
wire to his specifications. This was the beginning of the steel 
piano wire division of the American Steel & Wire Company, 
which today supplies practically all the piano wire used in 
the fine American pianos. 

The oldest record of the harpsichord is found in Eberhard 
Cersne's "Rules of the Minnesingers," published in 1404, 
No doubt such instruments were strung with the drawn iron 
wire from Germany, or possibly with softer wire, for such 
instruments were plucked with quills and were not struck 
by hammers, as were the pianos which came later. This was 
a little over three hundred years before Cristofori of Italy 
made his first pianoforte, in 1709. Then it was that drawn 


wire was put to some severe tests in tensile strength. The 
piano makers had difficulty because of breaking, for the 
stress on the string when struck by the hammers was so great 
that the strings often broke. The best modern hard-drawn 
wrought-iron wire can withstand only 70,000 pounds of pull 
per square inch, and this old German iron wire doubtless was 
much inferior. The modern steel wire used in today's pianos 
can withstand a pull up to 375,000 pounds per square inch. 

When the first pianos were built, however, the strings were 
not subjected to such great tension as they are today, and 
for a very good reason. The frames in which the wires were 
strung were first made of wood, and the tension could not 
be great. The builders found that the maximum stress on a 
wood frame was ten tons. Between 1770 and 1820 various 
wood frames were designed, but the best of them were un- 
satisfactory and had to be abandoned. Such frames were not 
sufficiently rigid to keep the strings in tune. In the upper 
registers, where the stress was greatest, special trouble was 
encountered, and even in lower registers the strings could not 
be kept from going flat. The frames also pulled out of shape 
and caused the key action to bind and stick. 

The first iron used in a piano frame was introduced by 
Joseph Smith of London in 1799, but the first all-metal frame 
was developed and patented by the English firm of Allen & 
Thorn, in 1820. This consisted of brass and iron plates in 
combination with metal tubes. Alpheus Babcock, an Amer- 
ican, patented a single piece iron frame in 1825, and Jonas 
Chickering improved upon this design in 1837. In 1859 Stein- 
way designed a cast-iron frame and the overstrung grand 
piano. Early, therefore, the iron frame became character- 
istic of American-made pianos. 


With the development of the stronger iron frame for the 
piano, the tension of the strings was increased. By 1862 piano 
frames could be subjected to a stress of sixteen tons, and 
today's pianos must withstand a terrific stress up to thirty 
tons. It is obvious that the old iron wire could not be used 
for such pianos, but along with the development of the metal 
frame, in fact the cause for the stronger metal frame, came 
the improvement of wire. Metallurgists developed new 
formulas for steel wire, and the steel mills devised new meth- 
ods of drawing and tempering it. The size of the wire was 
reduced, and the tensile strength was increased. The brilliant, 
rich tone of today's pianos is due to the development of this 
better wire, for after all, the heart of the piano is the strings. 
Music historians trace the piano to the primitive mono- 
chord, used by the Greeks. This instrument consisted of a 
single string which was made either of "catgut" or vegetable 
fiber. Its musical possibilities naturally were extremely 
limited; as a matter of fact, it was used not so much as a 
musical instrument as a scientific device for demonstrating 
the harmonic intervals of the musical scale. By dividing the 
string exactly in half, they showed how the octave was 
formed; by dividing it into thirds they demonstrated the 
octave and a fifth; by dividing it into fourths, the double 
octave; and so forth. But for some strange reason it never 
occurred to the gifted Greeks that they could make an instru- 
ment of many strings which would have the great musical 
resources of the clavichord or the harpsichord. 

Possibly one reason for their backwardness was a certain 
taboo against changing the number of strings found on their 
ancient lyres and harps. Writing about 200 A.D., the Greek 
Athenaeus recites a tale about a contemporary musician 


named Timotheus who had the temerity to increase the 
number of strings on the lyre. He says: 

"Whereas Timotheus the musician, coming to our city, 
has deformed the majesty of our ancient music, and despis- 
ing the lyre of seven strings, has by the introduction of a 
multiplicity of notes corrupted the ears of our youth, and 
by the number of his strings and the strangeness of his 
melody has given to our music an effeminate and artificial 
dress instead of the plain and orderly one in which it has 
hitherto appeared . . . The kings and the Ephori have 
therefore resolved to pass censure on Timotheus for these 
things, and further to oblige him to cut off all the super- 
fluous strings of his eleven, and to banish him from our 
domain, that men may be warned for the future not to 
introduce into Sparta any unbecoming customs/' 

If such penalties were inflicted on Timotheus for increasing 
the number of strings of the lyre from seven to eleven, what 
terrible things might have happened to anyone who dared 
to build a clavichord, comprising two or three octaves! Over 
a thousand years transpired before any builder of musical 
instruments had the courage to invent the many-stringed 
clavichord and harpsichord. 

We do not know exactly when the clavichord and harpsi- 
chord were originated, but it must have been during the 
latter part of the fourteenth century. Jean de Muris, writing 
in 1323 about the musical instruments of his day, does not 
mention either the clayichord or the harpsichord, but when 
Eberhard Cersne set about writing the "Rules for Minne- 
singers" in 1404, he mentioned both of these ancient ances- 
tors of the piano. The oldest specimen which has come down 



to us is the clavichord now on exhibit in the Metropolitan 
Museum, which is dated 1537. 

The clavichord was played by a keyboard much like that 
of the modern piano, but the internal mechanism and the 
way the string was contacted were entirely different. When 
the key was pressed down, a lever beneath the string came 
up. On the end of the lever was a piece of brass about an 
inch high and about a quarter-inch wide. This brass wedge 
struck the string and set it vibrating. The sound was feeble 
and thin, however, and not much like that produced when 
the piano key is struck. Some instruments had two key- 
boards, one for playing the full length of the string, and 
another for playing the partial length of the string. In playing 
the latter keyboard, the wedge struck the string at a pre- 
determined place in the length of the string and divided it so 
that the pitch of the note was not that of the entire length 
of the string but only that part between the wedge and the 
end of the string near the keyboard. This device doubled the 
resources of the few strings. 

The clavichord was an intimate instrument, suitable for 
the home and for sweet melodies. At first it had but few 
strings. Praetorius, writing of musical instruments about 
1619, describes a clavichord with only twenty-two strings. 
Later the clavichord had a compass of three to four octaves. 
The great J. S. Bach was very fond of the clavichord and 
spent many hours at this instrument playing his own ex- 
quisite music and the compositions of others. His son Em- 
manuel expressed a definite preference for the clavichord 
over the harpsichord. Later Mozart learned to love the 
clavichord, and it is said he composed on this quaint instru- 
ment the music for his "Magic Flute." When he became 


acquainted with the piano, however, he dropped the clavi- 
chord in favor of the more resourceful piano. Beethoven also 
is on record as praising the beauty of the clavichord, although 
he sponsored the piano and did much by his playing and 
composing to establish this instrument. 

Although Mozart went from the clavichord to the piano, 
it is said he never mastered the latter instrument. He was 
unquestionably a great artist on the clavichord, for it seemed 
to suit his delicate touch and adapt itself to his delicate 
musical ideas. But playing the piano was different, and 
Mozart was never quite able to change over to the technique 
necessary for playing it. It was the custom in playing the 
clavichord to use only the three long fingers of each hand, 
the little finger and the thumb never touching the keys. The 
touch on the keys must be light and delicate, for the clavi- 
chord could not be struck vigorously as can the piano. The 
strings were not pulled tightly as on the piano, and they 
were set in vibration with the light touch of the brass wedge. 
Dynamics were not used, for the string could sound only one 
way, and that was softly and sweetly. Notes could be sus- 
tained only for a moment, each touch of the key producing 
its own sweet, brief, delicate tone. Music written expressly 
for the clavichord must have sounded exquisitely beautiful 
on this instrument. Much of its beauty unquestionably is 
lost when played on the piano, for the piano is not suited to 
the moods and colorings of clavichord music. It remained for 
Beethoven to recognize this fact, and he wrote a different 
kind of music expressly for the piano. And if clavichord 
music loses much of its charm when played on the piano, it 
is also true that today's piano music could not have been 
played at all on the clavichord. 


Another multiple-stringed instrument of the time was the 
harpsichord. It also had a keyboard like that of the piano, but 
otherwise the internal mechanism and the way the string 
was contacted were different. When the key was pressed 
down, a lever beneath the string was raised. On the end of 
the lever was a sort of wooden peg, called a jack, in which 
was inserted a quill pick, usually made from the wing feather 
of the crow. When the lever was raised, the quill was raised 
past the string, and as it passed it plucked the string and 
set it vibrating. 

There were several varieties of the harpsichord, A small 
one, often held on the lap, was called the virginal. It is so 
called from its use by young girls to accompany their own 
songs. One of these instruments is interestingly described 
by the music historian Virdung, writing in 151 1. A somewhat 
larger variety was called the spinet, from its resemblance in 
appearance to a spinet writing desk. We have records of such 
instruments dating back to 1490, although the oldest one 
extant is dated 1521. The spinet usually had one row of 
jacks and only one string for each note. Its compass was 
usually three to four octaves. 

The musical effect of the harpsichord and its smaller 
varieties was much like a plucked guitar or zither. It was 
essentially a plucked instrument, and there were certain 
limitations to its performance which can be expected of a 
plucked instrument. It could not be played loud and soft, 
as can the piano, for no matter how hard the key was pressed 
down, all that could happen was that the quill passed by the 
string and plucked it. Notes could not be sustained for long, 
and when the musician stopped striking the keys, the music 
stopped abruptly. This led to a form of composition adapted 

to the peculiarities of the instrument. Climaxes had to be 
built up by a great flood of notes and not in the pressure with 
which the notes were struck. To fill in the gaps between 
certain periods in the music, cadenzas, arpeggios and other 
embellishments were employed in profusion. Harpsichord 
music, therefore, was highly ornate and intricate. For this 
reason it became popular with the composers of polyphonic 
music, in which two or three melodies could be played simul- 
taneously against each other. Such music flowed from the 
harpsichord in a stream of intricately woven and closely 
knit harmony, without much interruption and, it must be 
said, without much to relieve the monotony except the in- 
terestingly worked out musical pattern. 

The harpsichord has had an illustrious history, from 1400 
until after 1800. This venerable instrument was present at 
the birth of both the opera and the oratorio. When Jacopo 
Peri in 1600 produced his "Euridice," the first opera ever 
presented in public, the harpsichord mingled its voice with 
the lutes and lyres and flutes, much to the pleasure of the 
Italians of Florence. And when Emilio del Cavaliere in 
1600 produced his "Representation of the Mind and Soul," 
the first oratorio ever heard, the harpsichord sang along with 
the double-guitar and other strange instruments. 

Having demonstrated its prowess in such early opera and 
oratorio, it became the principal instrument of the orchestra 
during the time of Bach and Handel. The conductor, who 
usually was also the composer, sat at the harpsichord and 
played and directed the music. The orchestra in those days 
was not as dependable and capable as it later became, and 
it often limped and faltered. The conductor was a sort of 
prompter, and when some instrumentalist or section got off 

pitch or lost the rhythm, the conductor at the harpsichord 
came to the rescue and carried the performance along. The 
various choirs were also unstable, and often certain musicians 
were not available. The important parts of the missing 
players could be played by the conductor on the harpsichord, 
and thus the gaps left by the missing players could be filled in. 

But the instrument was not simply a "troubleshooter," 
It had its own important part to play, aside from "pinch 
hitting" for the missing or faltering players. The early com- 
posers did not know how to write for orchestra in the full 
and complete way they later learned to do. A kind of short- 
hand transcription of the music was given to the harpsichord, 
called the figured bass, and this consisted of chords in a more 
or less rhythmic beat. This was a sort of transparent music 
which meant "all things to all men" in the orchestra. Guided 
by this general idea of the music, the individual players more 
or less improvised whatever came to their mind. With such 
an uncertain type of composing, it is no wonder the composer 
sat at the harpsichord and nursed the players along. Even 
when definite parts were written for each player, the orches- 
tration was so lacking in body and solidity that it was 
necessary for the harpsichord to furnish a substantial back- 
ground of chords and musical embellishment to hold the 
complete ensemble together. 

The conductors of those days had not yet learned how to 
stand in front of the orchestra and beat time so all the players 
could perform to the same tempo. They sat at the harpsi- 
chord and, by playing along and emphasizing the beat and 
by nodding their heads, they maintained the rhythm and 
beat. Sometimes they might lift one hand from the keyboard 
and emphasize the beat, while playing along with the other 


hand. Later, two harpsichords were used, one for the con- 
ductor in the center and another for the player who per- 
formed the figured-bass part. Next to the harpsichord stood 
the first chair violinist. The beat was given to this player, 
who passed it along to other members of his section and 
thence to the rest of the players. 

This method of conducting started in Italy, and when 
opera entered Germany the same custom accompanied it. 
Handel, who studied in Italy, brought the custom with him to 
Germany and later introduced it in England. Haydn directed 
from the harpsichord, but toward the close of his career he 
added a time beater who stood in front of the orchestra and 
beat time while Haydn played at his instrument. When 
Beethoven definitely espoused the piano in place of the 
harpsichord he often sat at the piano and played while 
another man stood in front of the orchestra and beat time. 
Some of the early conductors occasionally beat time with 
their hands or with a roll of music. Lully is said tp have 
beaten time by pounding on the floor with his cane. On one 
particularly frantic occasion, in an effort to keep the orches- 
tra up to tempo, he accidentally hit his foot with his cane 
and injured it so that later his leg had to be amputated, 
indirectly causing his death. It was not until Mendelssohn 
and Berlioz that the modern custom of directing with a baton, 
and without either harpsichord or piano, became common. 

It has been shown that neither the clavichord nor the 
harpsichord could play both soft and loud. This was ob- 
viously a distinct disadvantage. It was to overcome this 
limitation thatJBartolomeo Cristofori, a harpsichord maker 
1>f Florence, invented th?pia!tufurLe, in 1709. Our word piano 
is a contraction of "pianoforte," which means literally 


"soft-loud." The word was first used in a letter from an 
instrument maker, Paliarino, dated June 27, 1598, and ad- 
dressed to Alfonso II of Italy, but it is generally thought this 
instrument did not use hammers and was merely a clavichord 
or harpsichord with some sort of dynamic control for enabling 
the instrument to play both soft and loud. We know numer- 
ous attempts were rfiade to remedy this fault in the old 
clavichord and harpsichord, and this probably was an in- 
strument with one of these many devices. 

Cristofori's piano incorporated two essential features which 
really permitted the player to perform on the instrument 
both loudly and softly, and his name has been made immortal 
because of them. These were an ingenious arrangement of 
hammers which permitted an instant escapement after the 
string was struck, and strength in the piano body which 
could withstand the great stress of the stretched string. 
When the key of the keyboard was pushed down, the hammer 
struck thp* strings a sharp blow from below and instantly 
dropped back away from the string, permitting it to vibrate 
freely. This may sound like a simple thing to accomplish, 
but nobody before had been able to do it successfully. The 
arrangement of the mechanism for doing this reminds one a 
lot of some of .the famous cartoons of Rube Goldberg called 
"Useless Inventions." The key is pushed down; this acts 
on a lever with a pin in it; this releases another gadget which 
falls, tripping another device under spring tension; this in 
turn actuates another part; and so on. But the result is a 
sharp blow on the strings and instant escapement of the 
hammer from the string. This mechanism has been modified, 
simplified and improved in literally dozens of ways, but the 
essential idea is still that of the ingenious Cristofori. 


The stress on the string from the clavichord brass wedge 
or from the harpsichord quill is much less than that when the 
string is struck sharply by a hammer. To withstand this 
blow, it was necessary to supply several strings for some of 
the notes and to strengthen the frame of the piano. This was 
also an ingenious bit of mechanical engineering. How the 
stress was increased up to ten tons, sixteen tons and finally 
up to thirty tons in the modern piano has already been told, 
The troubles with soft wire strings, the harder iron strings 
and finally with the modern steel wire strings have also been 
recited. It is all a thrilling story of the working out of the 
modern piano. And although many persons have shared in 
the glory of developing the piano to its modern perfection, 
this does not diminish the credit ojf Cristofori who first made 
the piano possible. - 

Harpsichord makers of other nations attacked the same 
problem, working along the line first laid down by Cristofori 
or along new lines. The Frenchman, Marius, in 1716 experi- 
mented with the harpsichord in an effort to save the cost of 
the expensive crow quills by substituting hammers. He also 
tried various ways of striking the string from beneath, 
from above and even horizontally from the side. The first 
successful French piano, however, was rnade in 1777 by 
S. Erard, better known for his double-action harp. In Ger- 
many, Silbermann followed Cristofori and made some beauti- 
ful pianos, one of them being made for Frederick the Great. 
In 1747 Emmanuel Bach played on this instrument, and from 
his reports it must have been a most creditable one. In 
England, among the famous builders of pianos was Broad- 
wood, who contributed an important improvement by deter- 
mining the point on the string where the hammer should 


strike for best tonal quality. When any string is struck, the 
fundamental, or pitch note, sounds, but the overtones are 
also produced. Of these overtones the seventh and ninth 
are inharmonious with the tempered scale. Their presence 
in the complete tone causes an unpleasant dissonance. It 
had always been the custom to strike the string at its center, 
but Broadwood determined that if the string were struck 
at a point near where these two harmonics occurred, they 
would be dampened out. Ever after this important discovery 
the tone of the piano has been sweeter and more beautiful. 

As has been the history of almost all improvements in 
musical instruments, the piano was not accepted immedi- 
ately but had to battle the old clavichord and harpsichord 
for years before it gained precedence over them. Players who 
had mastered the technique of the clavichord and harpsi- 
chord did not want to labor to change their style of playing 
in order to be able to perform on the piano. There was a 
distinct difference in the touch and technique necessary for 
the piano, and those who played well on the old Instruments 
were willing to let well enough alone. Mozart, who changed 
from the clavichord and harpsichord to the piano, is said 
never to have quite adapted himself to the new instrument. 
If such a genius as Mozart had difficulty adapting himself 
to the new technique, we can imagine that others found It 
most difficult. 

Besides, the early pianos were full of faults. The great 
stress of the stretched strings pulled the frame out of shape 
and caused the mechanism to stick and bind. The strings 
themselves would not stay in tune, for the frame was not 
rigid enough to maintain a uniform tension. The upper regis- 
ter, where the stress was greatest, gave special trouble. 


Many musicians thought the piano was impractical and 
predicted its early demise. It was not until 1768, nearly sixty 
years after the piano was invented, that it was first used as 
a solo instrument in England. On this occasion, J. C. Bach, 
son of the great J. S. Bach, played the piano in London. 
No doubt this had its effect in inducing the conservative 
King's Own Band to adopt the piano instead of the old 
harpsichord in 1795. So little recognition did the piano have 
that there was no piano music published until 1770, when 
the great piano artist dementi published several sonatas for 
the piano. 

It was about this time, too, that Mozart definitely adopted 
the piano and used it for the next twenty years until he died. 
The most famous of Mozart's instruments was a superb 
five-octave instrument made by Anton Walter of Salzburg. 
It had black ebony natural keys and white ivory sharp 
keys, the opposite of the arrangement used today. This 
reversal of the black and white keys was not uncommon in 
Mozart's day and before, this being the arrangement of 
keys found on many clavichords and harpsichords as well. 

It is interesting to note that the fame of the piano had 
spread about this time to America. Thomas Jefferson, in 
1771, was on the point of buying a clavichord for his new 
bride but changed his order to a "fortepiano." He and his 
fiancee had often played together, he on the violin and she 
on the clavichord, and he planned to present her with a fine 
clavichord as a wedding present. But in a letter to England, 
dated June I, 1771, he says: 

"I must alter one article in the invoice. I wrote therein 
for a clavichord. I have since seen a Fortepiano and am 


charmed with it. Send me this instrument instead of the 
clavichord: let the case be of fine mahogany, solid, not 
veneer, the compass from Double G to F in alt, and plenty 
of spare strings; and the workmanship of the whole very 
handsome and worthy of the acceptance of a lady for 
whom I intend it." 

Haydn was one of the first to appreciate the peculiar 
advantages of the piano and to recognize the definite change 
in playing technique from that necessary on the clavichord 
and harpsichord, but it was Beethoven who established the 
piano in a firm position. He not only mastered the technique 
for playing the piano but saw that it was necessary to write 
a different kind of music if the piano was to be exploited 
fully. He realized that music written for the clavichord or 
harpsichord did not bring out the peculiar beauties of the 
piano. Just as composers came to realize that music written 
for the flute was not suitable for the clarinet or the trumpet, 
so Beethoven saw that the piano had a beauty and an 
eloquence which could not be given full expression when 
required to produce music written originally for the clavi- 
chord or harpsichord. He composed great, solid, sustained 
chords, shaded from diminuendo to crescendo, and wrote 
sparkling staccato passages which were unknown to the old 

The delicate-touch players gave way to the piano pound- 
ers. This is reflected in the action piano builders gave to the 
piano mechanism. To withstand this terrific assault upon 
the piano, the action was increased to three or four ounces* 
pressure on the end of the key at Middle C. The performer 
retaliated, as it were, by changing his playing position in 

golfing language, his stance. Instead of sitting low at the 
piano with the forearm level with the keyboard, as had the 
players of the old school, such men as Liszt sat high above 
the keyboard so they could pounce upon it with more fire 
and vigor. The school of Chopin and Rubinstein finally won 
out, however, and a more delicate touch was adopted. This 
change was again" reflected in the making of the piano 
mechanism, for the touch on the end of the key at Middle C 
was brought down to about two and a half ounces, where it 
remains today. 

The modern piano is a great instrument which rivals the 
orchestra itself. Its eighty-eight notes, comprising seven 
octaves and four semitones, match playing range with the 
orchestra, from the lowest note on the giant BBb bass tuba 
to the shrill top note of the piccolo. It can play both loud 
and soft and has a wealth of technique and effects which 
puts it in a class by itself. 


Ancient Greek and Egyptian mythology has picturesque 
stories about the discovery or invention of the harp. One of 
the best is about some god of music who happened to be 
walking along one fine day and kicked the shell of a tortoise. 
Across the shell were thin strips of flesh or skin which had 
dried taut. The blow from the foot set these stretched strings 
vibrating and produced a sweet musical effect! There are 
other stories told by other peoples, and the most plausible of 
these is that the harp was originally suggested by the bow. 
After the arrow was dispatched, the string continued to 
vibrate and produced a musical note. 

This latter story has much more to support it than the 



former. Pictures of Egyptian and other ancient harps 
which have come down to us closely resemble the bow, for 
they do not have the post of the more modern harp, They 
consist of a framework of wood, curved or shaped like two 
sides of a triangle and having strings from one side to the 
other. The conventional post of today's harp, parallel with 
the strings, is missing on all ancient harps. It was left for the 
Irish and Saxons of the ninth century A.D. to add this im- 
portant feature. Its superiority has recommended it to harp 
builders ever since and is a prominent feature of the modern 

All ancient harps were diatonic, and semitones were ob- 
tained by pressing the string with the finger. The Tyrolese 
improved somewhat on this method by screwing little crooks 
of metal into the neck, which could be turned against the 
string to shorten it. In 1720 the Bavarian Hochbrucker in- 
vented a harp with five foot pedals which instantaneously 
shortened the C, D, F, G and A strings a semitone. Although 
this invention pointed the way for future development of the 
harp, it was not successful because the stopped notes were 
not of good quality and the construction of the pedal mech- 
anism was so flimsy and frail that it gave much trouble 
to the player. M. Simon, a harp maker of Brussels, was so 
sure that foot pedals were impractical that about 1758 he 
built a harp which was made chromatic by seventy-eight 
strings. This is only ten less than the modern piano has, and 
it must have required a tremendous reach to play all these 
strings, to say nothing of being able to find the notes in this 
thicket of catgut. 

Cousineau, a Frenchman, decided he would outdo Hoch- 
brucker and built a harp with fourteen pedals. This doubtless 


simplified matters for the hands but greatly complicated 
matters for the feet! It was a countryman of Cousineau who 
finally solved the riddle in 1810. S. Erard began experiment- 
ing with harps in 1786, and after working at the problem for 
nearly a quarter of a century he brought out his famous 
double-action harp which forms the basis for today's harp 
and did for the harp what Cristofori did for the piano. 

His harp was equipped with seven foot pedals, and the 
range was increased to six and a half octaves. These seven 
pedals acted on the following strings: Gb, Db, Ab, Eb, Bb, F 
and C. The clever part of the invention was that when the 
pedal was pushed part way down, the strings were raised a 
semitone, but when pushed down all the way, the strings 
were raised a whole tone. Instead of the fourteen pedals of 
Cousineau, Erard secured the same results with half that 
number. This made the harp chromatic in all keys and 
brought about a great impetus to harp playing. 

The harp has the honor of sitting in with a lot of other 
strange instruments in the earliest ensembles out of which 
eventually grew the orchestra. When Balthasarini produced 
his "Ballet Comique de la Reine" in Paris in 1581, the harp 
was selected as one of the instruments. Eight years later the 
Italian Caccini employed the harp, along with viols, lutes 
and lyres, to play the music for his "Intermezzo." Among 
the instruments used by Monteverde in his "Orfeo," pro- 
duced in 1608, was a "double harp " with two rows of strings. 

It seems the Germans did not take to the harp readily, 
for Bach did not use it, and Handel, although he experi- 
mented with the harp in "Julius Caesar" in 1713 and in 
the "Esther" oratorio in 1720, seldom, if ever, used it 
afterward. Haydn omits the harp from his classical orches- 



tra, and Beethoven's lack of interest in the harp is evidenced 
by the fact that he uses it only once, in "Prometheus." The 
Germans before Beethoven can scarcely be criticized for not 
using the harp, for it was not fully developed in their day, 
but Beethoven wrote after Erard had brought the harp to a 
high state of perfection. Spohr, a somewhat later German 
composer, wrote much for the harp, but his wife was a 
harpist! Later Germans, such as Wagner, learned to use the 
harp effectively, but usually in some historical role to repre- 
sent the ancient lyre, as in "Tannhauser." 

If the Germans disliked the harp, the French composers 
seemed partial to it. Berlioz uses the harp often, and in 
"Faust" he runs true to form in going to extremes in any- 
thing he likes, by writing for ten harps. Meyerbeer wrote 
separate parts for two harps in "Le Prophete." Boi'eldieu, 
in his opera "La Dame Blanche," given in 1825, was the 
first to use harmonics on the harp; and Debussy, writing 
much later, in "La Her," uses harp harmonics with great 

Today's harp is a beautiful and resourceful instrument of 
forty-seven "catgut" strings, which, with the help of seven 
foot pedals, give it a chromatic range of six octaves and 
eight semitones, from Cb two octaves below the bass clef to 
G three octaves above the treble. The pedals, from left to 
right, act upon the following strings: Db, Cb, Bb, Eb, Fb, 
Gb and Ab. When the pedals are not in use, the instrument 
is in the key of Cb, and the flat signatures are best for the 
harp, since the strings are open and have better quality 
than the stopped strings. The pedals work substantially as 
did those which Erard invented. The movement of the pedal 
to the first notch lowers the strings a semitone, and the 

movement to the second notch lowers the strings a whole 
tone. The strings are plucked by the thumb and first three 
fingers, the little finger being too short to be of use. Normally 
the strings are plucked a little above the middle, but when 
plucked near the end a brighter tone, resembling the tone 
of the guitar, is produced. 

The harp has a wealth of technique and effects. Arpeggios, 
glissandi and chords are peculiarly suited to the harp. 
Staccato, tremolo and harmonics on the middle strings are 
employed in fact practically everything is playable on the 
harp except rapid chromatic passages, due to the necessity of 
changing many pedals. It is to overcome this shortcoming of 
the regular harp used in this country that the chromatic 
harp of Pleyel & Co. of Paris was patented in 1894. This 
harp is not used in America but is popular in Europe. It 
does not have pedals, but secures a chromatic scale by the 
use of two sets of strings, one set of white strings on the left 
for the natural notes and one set of black strings on the right 
for the sharps and flats. 


The old Greek kithara and the troubadour gittern became 
the Spanish guitar toward the end of the sixteenth century 
when some musicians of Spain changed it from a four-string 
to a five-string instrument. 

Music terms are used so carelessly in ancient literature 
that we are not at all sure what instruments are referred to. 
The old troubadours and minnesingers played lutes and lyres 
and harps and gitterns, but just how much any of these were 
like what we have come to call the guitar is shrouded in 
ambiguity. The first trustworthy evidence of the existence 



of the guitar as we know it today is a representation on the 
Gate of Glory of the church of Santiago de Compostella, in 
Spain. This gate was erected in 1188 A.D. and definitely 
establishes the existence of the guitar for nearly a thousand 
years. Whether it was native to Spain or whether it was 
imported into Spain is not known, although there is a widely 
accepted belief that the Moors brought the instrument to 
Spain a century or so earlier than 1188. There are records of 
a guitar-like instrument in Spain from the thirteenth to the 
sixteenth centuries. This was variously called a Spanish lute 
or vihuela. Instruction books and music were published for 
the vihuela by Luis Milan in 1535. 

The early Italian opera composers wrote for the guitar in 
the latter part of the sixteenth and in the seventeenth cen- 
turies. At least we can fairly assume this, for they called for 
a variety of "lutes," and among them must have been some 
guitars, for the guitar had for several hundreds of years been 
a popular instrument. Cavaliere, writing the music for his 
first oratorio, in 1600, definitely names a "double guitar " 
among the miscellaneous instruments of the "orchestra/' 
Monteverde calls for two "large guitars" in his "Orfeo," 
performed in 1608. 

Ferdinand Sor, a guitar virtuoso, brought the guitar from 
Spain to England, and it became so popular as to put the 
old English guitar, or zither, out of business. The story is 
also told that this wave of popularity almost swept several 
other instruments out of existence, too. The endangered 
instruments are variously said to have been the harp and 
the harpsichord. The fact that one story says the harp was 
endangered and the other says it was the harpsichord casts 
suspicion on the whole story. Anyway, the story goes that 


about 1750 the guitar became so popular as to stop the sales 
of the harp (or harpsichord). The harp (or harpsichord) 
manufacturers killed the craze by a clever move. They se- 
cured a large number of guitars and put them into the hands 
of people of the lower classes servant girls, stableboys and 
the like. This stamped them as instruments not worthy of 
persons of culture, and the guitar craze quickly died out. 

This stigma did not by any means cling to the guitar, for 
we find Rossini writing for the guitar in his "Barber of 
Seville," and Weber calls for it in his "Oberon." Berlioz, 
the great French composer and conductor, was a guitar 
player; in fact, this was the only instrument he could play 
at all, and we can see the influence of the guitar in the chord 
arrangements of his compositions. Paganini, the celebrated 
violinist, also was a proficient performer on the guitar. The 
most famous guitar virtuoso today is the Spaniard musician, 
Andres Segovia. 

Those who think of the guitar as an instrument fit only 
for strumming chords as an accompaniment to love songs 
should hear such guitar artists as Segovia play the guitar. 
In their hands it becomes an instrument of rich beauty and 
great musical resources. There are two ways of playing the 
guitar. The familiar chording is called "rasqueado." The 
great artists, however, use the "punteado" style of playing. 
This is a melodic style, and the possibilities of this kind of 
playing are limited only by the ability of the musician. 

Today's guitar has six strings: E, A, D, G, B and E from 
E below the bass clef to E above. Its sides and back are made 
of maple, ash, cherry or mahogany, while the top is made of 
deal or spruce. The neck, fingerboard and bridge are of 
rosewood or ebony. The old-style guitar is known for its 



round hole in the top, under the strings, but many of today's 
finest guitars are made with the ff holes of the violin. The 
tuning formerly was done with wooden pegs, such as are 
still found on the violin, but in 1790 the Germans introduced 
the metal screws, which are easier to operate. 


The banjo is native to America, being a creation of the 
American Negro. It is a four-string instrument incorporating 
the unique feature of a soundboard made by stretching skin 
over a hoop. The bridge rests on this taut skin, and the 
strings are stretched over this bridge. No doubt the rough 
idea was brought with the Negro from Africa, where it was 
picked up from Arabian traders who played the Arabian 
rebab or other similar string instrument. In slave days it 
was the most popular instrument in the South. Thomas 
Jefferson, writing in 1784, says it was known to Negroes as 
the "banger/ 5 Up to 1830 it was commonly known as the 
"bonja." Today this word has changed into "banjo." There 
are two types of banjo : the tenor and plectrum, the latter 
being less common. The former is strung as follows, in 
ascending order: Middle C, G, D and A. The latter is strung 
Middle C, G, B and D. 



How Music Is Made 


THE NEXT TIME you hear a band or orchestra play, ask 
yourself what the stuff is that you call music. WHAT is 
it when the band is playing? WHERE is it when the band 
stops? Because of its elusiveness we have got into the habit 
of regarding it as something unlike the ordinary physical 
things the world is full of. This is wrong. A little thought 
will show you that the stuff of which music is made is every- 
where around you. 

A man who had an automobile without a speedometer 
used to speak of "turning the corner at thirty miles an hour," 
"riding pleasantly along at twenty miles an hour," "opening 
her up until she was doing forty miles/' When asked how 
he could tell the speed at which he was traveling when he 
had no speedometer, he said, "Well, when she gets to going 
twenty miles an hour the right fender begins to rattle. When 
she gets to going thirty an hour a little bolt in the dash begins 
to jingle. When she hits forty the glass in the windshield 
chatters terribly." 

Rates of vibration! Not music, of course, but the stuff of 


which it is composed; the only difference is that music is 
more orderly and better controlled. 

An engineer once was called in to investigate a large 
centrifugal fan which was rapidly destroying a brick air 
stack. The fan ran just like an ordinary electric fan and was 
used to ventilate a certain factory, pulling the foul air from 
the rooms and sending it up this ventilating stack. When the 
fan was going full speed the brick stack would quiver and 
tremble and vibrate, so that the bricks began falling out. 
The fan was mounted on reinforced concrete, and seemingly 
all the vibration from the motor was eliminated, but this 
didn't remedy the vibration of the stack one bit. Finally this 
engineer, who had a knowledge of the laws of sound, was 
called in, and he solved the problem in a few minutes. 

Every time the fan went around, each of the blades in the 
fan boosted a column of air up the stack. When going at 
full speed these puffs of air went up the stack at regular 
intervals, causing a kind of "beat/ 5 It just happened that 
this stack was the right length to vibrate or "beat 5 * in 
sympathy with the blades of the fan. This caused a sym- 
pathetic pulsating in the stack that became stronger and 
stronger until the stack was ready to fly to pieces. 

There were two things that could be done to correct this 
evil. Either one of them would destroy this "beating to- 
gether" of the fan and the stack, and everything would then 
run smoothly. One was to shorten or lengthen the stack so 
the length of the air column would be out of sympathy with 
the speed of the fan. The other was to gear the fan either up 
or down so its vibration would be thrown out of sympathy 
with the length of the stack. It was decided to shorten the 
stack just a trifle. Presto! The riddle was solved. 



Music ? Not exactly, but some of the stuff of which music 
is made. 

A man who is now a prominent acoustical expert was once 
employed as a consulting engineer in one of the largest auto- 
mobile factories in the country. He had a hobby of experi- 
menting with sound, and his knowledge of the laws of sound 
once enabled him to solve a great engineering puzzle in a 
new model car that was being produced at the time. 

This particular car had a bad hum in it when traveling at 
a certain speed. As is usual in such circumstances, the gears 
were suspected, but days and days of experimenting failed 
to locate the trouble. This particular engineer stood out 
against the gear theory, but his associates would not listen 
to him. 

He discovered that at the rate at which the hum occurred, 
the rear wheels were making two revolutions a second. He 
also discovered that the hum had a definite pitch, closely in 
tune with F# below Middle C. That meant that this hum was 
being produced by a vibration of about 180 a second. 
Checking the teeth in all the gears suspected, he saw that 
none of them could produce this rate of vibration. 

This much proved, he began looking for the real trouble. 
He found it when he counted the knobs on the tire tread. 
There were 90 of these, and the wheels were making two 
revolutions a second. Somewhere in the car was a resonator, 
which reinforced this vibration of 180 per second and made 
it a strong hum. Simple enough! Rib-tread tires were sub- 
stituted, and the hum vanished. 

Hardly music such as the symphony orchestra turns out, 
anyone will admit, but you can recognize the relation. 

Some musicians playing a dance job at a summer resort 


used to relate how they tuned their instruments with an 
electric fan. Near the stand an ordinary electric fan was 
running, and its hum was recognized as a distinct note. 
These musicians finally determined that this note was 
pitched exactly an octave below the "pitch" note of their 
Bb trumpets and exactly with that of the Bb trombones, 
By playing their trumpets an octave higher and their trom- 
bones in unison, they were able to obtain the correct pitch 
for their instruments. It just happened that the blades of 
this fan struck the air at the rate of 116.5 times per second, 
the vibration rate of the Bb in the second octave below 
Middle C. 

Not exactly what would be called dance music, but made 
up of the same raw material. 

The reality of the stuff of which music is made is strikingly 
illustrated by the Illinois School for the Deaf Boys' Band of 
Jacksonville, Illinois. These boys cannot hear themselves 
play, and yet they play in unison and with a feeling and 
appreciation that are the envy of many bands that can 

Their first music lesson was through the use of a baseball 
bat. The bat was used to count out rhythm on the floor. 
The vibration was transmitted through the floor to their 
feet, and they learned to march to the music of the ball bat, 
first in their bare feet, later with their shoes on. Drums were 
then secured, and they were able to sense a more refined 
music. After some training they could rest their hands on a 
piano that was being played and "hear" the music through 
their fingertips. Band instruments were finally purchased, 
and now their sense of "hearing" through the nerves of 
their bodies is so acute and so well trained that they are 


recognized as a first-class band, able to hold their own with 
other bands that have normal sense of hearing. 

The sound wave which seems so beneficent and beautiful 
can also become a terror when it gets out of control, because 
of the tremendous forces wrapped up in it. This is illustrated 
in the ventilating stack. It is also illustrated in incorrectly 
designed auditoriums where the echo or reverberation be- 
comes so violent as to cause the auditorium to tremble from 
top to bottom. 

The story is told of a violinist who wagered he could 
destroy a wagon bridge merely by playing his violin. He was 
laughed at and told to go as far as he liked. He chose a note 
on his violin that he found vibrated in unison with the bridge, 
and began to play. It is said that the bridge began to vibrate, 
and the longer the same note was sounded on the violin, 
the more violently the bridge rocked. The authorities of the 
town at last became so alarmed that they compelled him 
to cease, for fear the bridge would break to pieces. 

There is a basis of fact in this story. It is well known that 
soldiers are required to break step when marching across a 
bridge, because the rhythmical step of marching men sets 
up a pulsation in a bridge which becomes stronger and 
stronger and threatens to rock it to pieces. A small dog 
trotting rhythmically across a bridge will shake a bridge 
to a greater extent than a horse which walks across with 
broken step. 


Now that you've been introduced to the raw material of 
which music is made, you'll want to become more intimately 
acquainted with the vibrating sound wave. 



Just like a zoologist when he finds a bug, well take the 
sound wave into the laboratory, dissect it and see what it is 
like. Of course, we can't see the sound wave, but we can learn 
something about it from observing the effects its activity 
brings about. 

Extremely delicate and sensitive machines have been in- 
vented which have shown us what the sound wave is like. 
In the first place it is known that a sound of any pitch has a 
fixed number of vibrations per second. Middle C on the 
piano vibrates 261.6 times in a second. The C which is three 
octaves below it vibrates only 32.7 times per second, while the 
C three octaves above vibrates 2,093 times per second. 

There is nothing particularly mysterious about vibration 
or the rate of vibration. Did you ever catch a weed or a stick 
in your bicycle which played a tattoo on the spokes of the 
wheel ? The faster you rode, the higher the pitch of the note 
would rise. There you have in a crude illustration what rate 
of vibration is. 

Men who study sound waves sometimes fasten a piece of 
cardboard so it will rest against the pegs of a revolving wheel, 
such as the wheels of fortune you see at carnivals and recrea- 
tion parks. As the wheel turns, the pegs "tick" the card- 
board. A note is produced in this way, and its pitch is de- 
termined by the number of pegs that "tick" the cardboard 
in one second. 

In studying anything it is helpful to look at a cross section 
of it, and it is easy to get a cross-section view of a tone wave. 
The following illustration is not entirely satisfactory, but 
it will help a lot in understanding what a tone wave is like: 
The next time you ride a train, listen particularly to the 
sound of the bells that warn people at the crossing of the 

train's approach. As you zip past you will notice that the 
pitch of the bell starts high and takes a decided dip down- 
ward. Sound travels in dry air 1,132 feet per second at 70 
degrees Fahrenheit, and when you are close, the sound waves 
have only a few feet to travel to reach you. Consequently 
they reach your ear with approximately the same rapidity as 
they occur in the bell. As a matter of fact, when approaching 
the bell, the sound waves strike the ear more rapidly than 
they occur in the bell, and the pitch of the bell actually rises 
as you approach. But as you draw rapidly away from the 
bell they must travel farther, each vibration farther than 
the one preceding it. Therefore each one reaches the ear a 
little farther apart. The effect is that of having the vibrations 
start out rapidly and gradually become slower and slower. 
The result is a cross section of the sound which illustrates 
that the rapidity of vibration determines the pitch of a note. 
You notice the same thing when automobiles honk while 
passing each other at a high speed. 

There are rates of vibration below and above a certain 
number per second that we cannot hear. It is difficult to 
hear, as a continuous sound, vibrations of less than 16 per 
second, just as it is difficult to see, as a continuous moving 
picture, less than 16 exposures per second of a moving- 
picture film. The average silent moving-picture camera is 
run so 16 exposures are shown per second, and when less 
than that are shown per second the picture "flickers." So it 
is with low rates of vibration. Instead of the vibrations 
sounding as one tone, they separate into 16 separate pul- 

Some persons can distinguish sounds with a vibration 
rate as high as 20,000 per second, but the extreme for most 



people is less than that. The piano has a range between 27.5 
and 4,186 per second, while the pipe organ goes as low as 16, 
and some organs have a pipe which vibrates 8 times in a 
second, but this can hardly be called a musical note. In the 
orchestra, the lowest Bt> of the contrabass tuba has a fre- 
quency of 29.2 vibrations per second, while the top note on 
the piccolo is C, four octaves above Middle C> and has a 
frequency of 4,186. A good Turkish cymbal has a high fre- 
quency of 12,000 to 13,000 vibrations, and there are over- 
tones in the orchestra which are heard up to 12,000 or 15,000 
frequencies, but the musical scale is generally considered to 
lie between 16 and 4,186 vibrations per second. 

Music heard over the radio, from sound films, and on 
phonograph records is limited to a more restricted range. 
Few radio stations are equipped to transmit frequencies over 
10,000, and most are limited to 5,000. The lower range 
seldom goes below 100. Radio speakers generally are in- 
capable of picking up even these limits, most music on the 
radio speaker lying between 100 and 3,500 frequencies. It 
is obvious that the fundamental of the low-voiced instru- 
ments and most of the high harmonics are not heard at all. 
The same limits apply generally to music reproduced 
from sound film and heard in the moving-picture theater. 
Phonograph-record recordings pick up a greater range of 
frequencies, the best of them from 30 to around 10,000, but 
much of this is sacrificed on poor reproducing machines. 
"Canned" music, therefore, still has serious limitations. 

It may seem like an impossible job to count so many vi- 
brations per second. It isn't done exactly that way- The 
counter of vibrations starts at the other end. He invents a 
machine that is geared up so he can produce an exact number 


of vibrations per second, and then he compares the note to be 
determined with the note of known vibrations made by the 
machine. When they coincide, he knows the vibration rate 
of the note in question. 

These machines are capable of setting up vibrations as 
rapid as 500,000 per second, but those above 20,000 are not 
heard. The reason for this is that the human ear is not sensi- 
tive enough to pick these up and transmit them to the brain. 
Hearing results when an object outside the ear sets up a 
vibration whose pulsating is transmitted through the air to 
the ear and on to the brain. These pulsating air waves strike 
the eardrum, and this sets the eardrum vibrating. The nerves 
of the ear transmit this sensation to the brain, and hearing 
is said to result. 

It is easy to understand that vibrations can become so 
rapid that the eardrum membrane cannot vibrate as rapidly. 
When this point is reached we cease to hear. Some insects 
have an "ear" which is more sensitive than the human ear, 
and science has discovered that they use as calls and "lan- 
guage " certain sounds which are above the limits of sound 
for the human sense of hearing. 


Prize fighters know that they can get in many light licks 
while they are placing one heavy blow. They depend more 
on the quick jab than on the blow which must start a distance 
back and gain the momentum necessary to give a heavier 
jolt. While the long drive is on its way the other fellow may 
get in a couple of short, vicious jabs, and the "haymaker" 
may never land. Many short blows may be struck with 
greater effect than a few long ones. 



Likewise the rapidly vibrating sound wave is not as long 
as the one that vibrates more slowly. We can, therefore, 
speak of the note of low pitch as being of longer wave length 
than the note of high pitch. When a string is stretched and 
plucked, the notes become higher as the length of the string 
is shortened. The high notes on the piano and Italian harp 
are made on the strings that are short, and the low notes on 
the long strings. 

Yes, you say, but why is it that the G string and the E 
string on the violin are the same length while the notes they 
produce are widely separated in pitch? And why is it that 
some wires on the piano are the same length, yet they play 
notes of different pitch ? 

There is a point involved here that many do not under- 
stand and that has caused a great deal of confusion. To avoid 
this confusion you need only make the distinction clearly 
between the length of the string and the length of the vibrat- 
ing sound wave the string generates. The low bass strings on 
the piano are the same length, it is true, but those that 
generate the lower notes are wound with wire and weighted 
down so they cannot vibrate so fast. The same is true on the 
violin: the E string is small while the G string is either gut of 
large size or wound with fine wire to weight it down. The 
length of the sound wave must be distinguished from the 
length of the string or whatever it is that creates the sound 
wave. It is the length of the sound wave that determines 
the pitch, and the shorter the wave, the higher is the pitch. 

The tone waves of the musical scale, which we have already 
learned vary in frequency between 16 and 4,186 vibrations 
per second, vary also in length from about 70 feet to 3 inches. 
Middle C is over 52 inches long. 


The longest pipe of most pipe organs is over 32 feet long 
although there are a few organs with a pipe over 64 feet long. 
The 32-foot pipe produces a note which vibrates 16 times per 
second, and the tone wave is about 70 feet long. A tiny pipe 
has been made which is only a quarter-inch long and which 
has a vibration rate of about 16,000 per second, while the 
wave length is about a half-inch long. 

The cornet in Bb sounds the Bb in the second octave below 
Middle C as its fundamental note and the total length of its 
tubing is about 53 inches, while the tone wave is about twice 
this length. The trombone sounds as its fundamental note 
the Bb an octave lower and since this tone wave is twice as 
long as that of the octave above, we find the trombone just 
about twice as long as the cornet, or about 107 inches. The 
BBb bass plays an octave below the trombone, and we find 
it about twice as long as the trombone and four times the 
length of the cornet, or about 216 inches. The wave length 
of its lowest note is about twice this. 

C Piccolo 

About I 2" long Tone wave about 26" long 

C Flute 

About 24" long Tone wave about 52" long 

Bb Cornet 

About 53" long Tone wave about 117" long or nearly 10 feet 

Bb Trombone - _ 

About 107" long Tone wave about 235" long or nearly 20 feet 

BBb Sousaphone _ 

About 216" long Tone wave about 469" long or nearly 40 feet 

Fig I. 

Above are shown lengths of various instruments and the lengths 
of their tone waves. Octaves are exact multiples of each other, and 
the instruments are nearly so. 




Possibly you have noticed that the large G string on a 
violin plays a lower pitch note than the E string, that it seems 
to vibrate farther from side to side than the E string and 
sounds louder. You also will notice that, as the length of the 
G string is shortened by pressing the string down against the 
fingerboard, the note is raised in pitch, the string vibrates a 
shorter distance from side to side and loses some of its loud- 

If you have observed this you have observed another factor 
in how music is made. Not only does the number of vibra- 
tions enter into the making of music, but the width of the 
vibration as it travels from one side to the other also has 
something to do with it. "Loudness" is the chief quality due 
to the width of sound waves. 

However, it is not wholly responsible for loudness, for 
loudness is made up of both rate of vibration and width of 
vibrating wave. A high note can be made to sound softly and 
a low note loudly. It is possible to play a low note loudly 
enough to be heard as far away as a high note played 

It is similar to shooting a pea with great velocity and 
tossing a larger pebble easily. When they strike a scale the 
beam flies up equally high in each case. Rate of speed in the 
pea makes up for lack of weight, just as rate of vibration in 
the tone wave makes up for lack of width. Weight in the 
pebble can also make up for lack of speed, just as width of 
tone wave can make up for lack of rapidity of vibration. 

The boom of a tympano and the boom of a cannon may 
sound the same pitch note, but the boom of the cannon will 


carry farther because the width of the vibration is larger. 
The African signal drums used by the natives are made from 
large logs and sound a deep bass note. Signals from these 
drums carry for miles. Foghorns on ships are pitched low 
because the sound will carry farther than a high-pitched 
note. On the other hand, in the band the oboe, which is a 
high-pitched instrument, can be heard from a distance, stand- 
ing out above all the lower instruments. The bell lyra or 
glockenspiel used in outdoor bands can be heard distinctly 
above the rest of the band. When bands are broadcast over 
the radio, as at football games, the bright "ping" of the bell 
lyra can be heard more distinctly than any other instrument 
in the band. This instrument is made up of short bars of steel 
which are struck with a mallet. It is high in pitch and seems 
to have some of the "loudness" qualities of the high-pitched 

The explanation of this "loudness" of the low-pitched 
drum or foghorn on the one hand and the high-pitched oboe 
or bell lyra on the other is a factor known as intensity. The 
width or amplitude of the wave of the low note may be great, 
but the number of vibrations per second may be small The 
total intensity of the note depends upon the amplitude of 
the wave as well as the number of vibrations per second. 
Likewise, the amplitude of the wave of the high note may 
be small, but the number of vibrations may be large. The 
intensity of the note, generally referred to as "loudness," 
depends upon amplitude of the wave and the number of vi- 
brations. Thus, the intensity of both the low and the high 
notes may be the same. 

"Loudness" is a psychological reaction to sound and is 
inaccurate in measuring intensity because such reaction is 



purely subjective. The acoustical engineer talks objectively 
of intensity, but to the layman this means merely "loudness." 


We have taken some measurements of the tone wave, 
having learned about its rates of vibration, its lengths and 
its widths. It is somewhat more difficult to understand just 
what the structure of the tone wave is like. 

One character of the tone wave can be illustrated by a 
train of cars. Well say the train has a hundred cars and has 
stopped on a sidetrack. The brakeman waves the engineer 
to back up, the engineer throws the reverse lever, and the 
engine backs. The engine hits the coal tender, the tender hits 
the first car, the first car bumps the second, and so on until 
this force is passed to every car and the whole train is in 
motion backward. In the same manner a vibrating object 
sets the molecules of air nearest to it in motion and each 
bumps into its neighbor. The vibration is passed on and 
on at the rate of 1,132 feet a second, the rate at which sound 
travels at 70 degrees Fahrenheit. 

It is misleading to say that the molecules move on and on: 
they do not move far, although the wave of energy may travel 
many feet or even miles. Each molecule moves with the wave 
until it strikes a neighboring molecule, when it stops and 
returns finally to its original position. The neighboring mole- 
cule passes the energy to its neighbor, and so on. It is some- 
thing like a relay race in which each runner stops when he 
tags the next man. 

If we cover the surface of a quiet pool with specks of 
powder and drop a rock into the water, a circular wave will 
spread over the pool, traveling outward from where the rock 


hits. The wave travels on and on, but the specks of powder 
do not move with it. They take the shape of the wave. In fact, 
it is only the shape of the wave that does travel. 

The manner in which a tone wave travels is also illustrated 
by a wheat field when a breeze passes over it. The heads of 
wheat are affected by the wave, bunching up and swaying 
forward as the wave passes over them; but they remain 
rooted in the ground, while the wave may travel miles across 
the field. 

It is also important to distinguish between movement of 
the molecules in the air and such movement of air as we have 
when the wind blows. As already mentioned, sound travels 
1,132 feet per second at 70 degrees Fahrenheit. This speed is 
at the rate of a mile in 4.7 seconds, or about the speed of a 
rifle bullet. It is also nearly 800 miles per hour. The most 
violent hurricane travels less than one tenth as fast, so we 
can see that it is not the air as a whole which travels, but that 
the impulse is passed along from molecule to molecule at this 
rate of speed. It is the flexibility or resiliency of air which 
makes this possible. Water is more resilient than air, sound 
in water traveling at the rate of about 4,800 feet per second. 
Steel is still more resilient, sound traveling in steel at the rate 
of 16,500 feet per second. 

Stick a pin into some soft wood, pull the end over (C, 
Fig. 2) and let it flip back. You get a musical note, and you 
can see the pin vibrate rapidly back and forth. When you 
pull the pin back and release the top, it springs forward just 
about twice the distance you had it drawn back. It rushes 
past its upright position (B, Fig. 2) and goes over to the other 
extreme, where it comes to a position of momentary rest 
(A, Fig. 2) and then comes back again (C, Fig. 2). This per- 


formance is a complete vibration, or cycle, and is the unit 
which goes to make up the tone wave. 

Fig. 2 

As it makes its first trip across, it compresses the tiny mole- 
cules of air together. The nearer they are to the pin, the more 
they are compressed. Out a distance in the air they are not so 
compressed, and still farther out they have not felt the move- 
ment of the pin and are their natural distance apart. But 
they will feel it when this area of compression reaches them. 

When the pin reaches the end of its first sweep (A, Fig. 3) 
it springs back to approximately its original position (C, 
Fig. 3). There is immediately a rushing backward of the 
molecules. They have been tightly pressed together, but 
when the pin reverses its motion it leaves an open space in 
its wake. 

Immediately next to the pin there are very few molecules, 
and they are far apart (C, Fig. 3). This area where the mole- 
cules are far apart is called a rarefaction. Between the pin 
and the area of compression the molecules are distributed 
with various distances between them. The farthest apart are 




b ' 

UJ & 





s \ 



i * 

a ^ \ 



5 g 





2 * 







LU . 
* * 


w III 


2 * 


' B 


ly A 








O I w 


2 >ui 
in / 5S 




{3 ^ 

2 i 



















la w 



ft *0 






IU ^ 



Z / 

* * 




near the pin, and the closest together are those in the area 
of compression. When the pin has sprung back on its return 
trip as far as it can go (C, Fig. 3) and starts forward on its 
second trip, the molecules begin to pile up against the pin 
again, and another area of compression is started (A, Fig. 3). 

In the complete tone wave we have the following: First, 
an area in which the molecules are compressed together, or 
a compression. Second, an area in which the molecules are 
their natural distance apart. Third, an area in which the 
molecules are widely separated, or a rarefaction. Fourth, 
an area in which the molecules are their natural distance 

We may measure the length of a wave from any point to 
the corresponding point farther on, but it is more customary 
to measure it from compression to compression. 

As the pin continues to vibrate, these complete tone waves 
are sent out. Each wave travels at the rate of 1,132 feet a 
second at 70 degrees Fahrenheit, and at the end of a second 
the first wave will be 1,132 feet from the pin. Between this 
wave and the pin will be quite a number of other waves com- 
ing along, the number depending on the rate of vibration. If 
the pin vibrates 261 times a second, it will sound like Middle 
C on the piano, and there will be 261 waves distributed along 
the 1,132 feet, and each wave will be equal to 1,132 divided 
by 261, or a little longer than 52 inches. Figure it out it's 
plain arithmetic. 

If a longer pin of the same thickness were used, it would 
not, of course, vibrate as rapidly. If it made only 32.7 com- 
plete vibrations per second, there would be only 32.7 tone 
waves along the 1,132 feet at the end of one second, and each 
would be nearly 35 feet long. Figure it out: 1,132 divided 

by 32.7 equals nearly 35. This is the length of the C which 
is three octaves below Middle C. 

These computations are made at a temperature of 70 
degrees Fahrenheit. If the temperature is warmer, sound 
travels faster, and if the temperature is colder, sound travels 
slower, than 1,132 feet per second. When sound travels faster 
than 1,132 feet per second the pitch of the note becomes 
sharp, for in one second there are more waves distributed 
along the 1,132 feet, and the more waves there are, the higher 
the pitch. Conversely, when sound travels slower than 1,132 
feet per second, the pitch of the note becomes flat, for in one 
second there are fewer waves distributed along the 1,132 
feet, and the fewer waves there are, the lower the pitch. 
That is why bands outdoors on a cold day sometimes sound 
flat, and why on a very hot day they sometimes sound 

This fact should not be confused with changes of string 
instruments under changes of temperature. The strings of 
the piano contract and become shorter, or rather tighter, 
when they become cold, and hence the pitch is raised. Con- 
versely, in hot weather they expand and become longer, or 
rather looser, and the pitch is lowered. This is exactly the 
opposite of the effect of temperature on wind instruments, 
That is why wind instruments and piano, when played to- 
gether, are so difficult to put in pitch when the weather 
varies. In cold weather the wind instruments go flat while 
the piano goes sharp; and in hot weather the wind instru- 
ments go sharp and the piano goes flat. In the finest radio 
and recording studios the temperature is held constant to 
avoid these difficulties. 



Our illustration of the train and its cars may be misleading 
on one point, and now is the time to correct this. Sound 
waves do not progress in a straight line only, but they radiate 
in all directions. This bumping of one molecule into its neigh- 
bor goes outward in all directions, like the radii in a sphere. 

When the vibrating object is the lips of a player and his 
lips are pressed against the mouthpiece of a horn, this sound 
radiates in all directions, but that part of the total wave that 
comprises the air column of the horn is made to sound louder 
through resonance. This principle of resonance is funda- 
mental in the horn, and it is necessary to understand what 
resonance is before one can understand how a horn makes 

This principle of resonance can be illustrated by a tube 
with one end in water and a vibrating pin held near the open 
end. By moving the tube up and down in the water, a spot 
will be found where the note sounded by the pin will become 
louder. When the tube is lowered or raised beyond this point, 
the sound will die away, but as long as it is held at this point 
the note will be heard louder. 

If the tube is held at this point, and another pin that 
sounds another note is held at the open end, the note of the 
pin will not sound louder. In order to make the note of the 
new pin sound louder, it will be necessary to move the tube 
up or down until another point is located. If the note of the 
new pin is lower than the first, the tube must be drawn up so 
more of it is out of the water. If the note of the new pin is 
higher than the first, the tube must be lowered so less of it 
extends out of the water. 

It is clear that a certain length of tubing is necessary to 
reinforce each note. This principle is illustrated on the 
marimba. The horizontal bars give comparatively little 


Fig. 4 

volume of sound, but when tubes of various lengths are 
placed below the bars, this small volume of sound is built up 
in the tubes, making each note sound loudly when its corre- 
sponding bar is struck. On a marimba these tubes, or reso- 
nators, are long for the low notes and gradually decrease in 
length as the notes become higher. 

What happens when a tone is reinforced and built up? 
We have already seen that as the pin vibrates it sends out 
waves one after another, one complete wave to each com- 
plete vibration. When the vibrating pin is held over a tube 


and moves from C to B (Fig. 4) a compression is started down 
the tube. As the pin moves from C to B, it gains momentum, 
for the pin swings like a pendulum, and a pendulum swings 
with greatest speed in the center of its arc. The faster the pin 
swings, the more compactly the molecules are pressed to- 
gether; therefore the molecules are closer together at B than 
near C. As the pin nears position A, it slows up and finally 
comes to momentary rest at A before beginning its return 
trip. As the pin swings in this manner from C to A it sends a 
compression down the tube. If we could see this compression 
we would see its center packed closely with molecules and 
its head and tail less closely packed, the degree of close- 
ness diminishing from center to either head or tail (see 

Fig. 3). 

By the time the pin comes to momentary rest at A, the 
head of the compression has traveled to the bottom of the 
tube, is reflected to the open end and is ready to escape into 
the air on its way to our ear. But as it escapes, so many of 
the molecules leave the tube that a scarcity of molecules is 
found in the tube and a rarefaction is created at the open 
end. This rarefaction begins just at the time the pin begins 
its return swing from A to B, which movement of the pin 
also starts a rarefaction. Two rarefactions, therefore, are 
created simultaneously, one by the escaping compression 
and another by the movement of the pin away from the open 
end of the tube. These two rarefactions join forces and be- 
come a larger rarefaction than either alone. 

It would be interesting to see what this rarefaction looks 
like. If our eyes were capable of seeing the movement of the 
molecules we would see exactly a reverse of the compression. 
The center would be comprised of only a few molecules 

widely separated, while the head and tail would be comprised 
of molecules less widely separated (see Fig. 3). 

It would also be interesting to see the working together of 
the two rarefactions. As the head of the compression starts 
to emerge from the open end of the tube, the head of a rare- 
faction starts to form and descend into the tube. As men- 
tioned above, the density of the molecules increases from 
the head of the compression to its center and decreases from 
the center to the tail. Therefore, as the head and middle of 
the compression emerge, an increasing number of molecules 
escape, thereby creating a correspondingly greater scarcity 
of molecules in the tube. By the time the center of the com- 
pression has emerged from the tube, the center of the rare- 
faction has been created. At the same time, the pin has moved 
from A to B, away from the tube. At B the pin is traveling 
at its greatest speed and is thereby creating a maximum of 

The next half of the performance is the reverse of the fore- 
going. As the pin moves from B to C, it gradually decreases 
its speed, and the degree of rarefaction gradually decreases 
until the pin comes to momentary rest at C. During this 
same interval, the second half of the compression escapes. 
Both the swing of the pin from B to C and the escape of the 
second half of the compression create the second half of the 
rarefaction. The escaping compression and the movement 
of the pin from A to C each creates its own rarefaction, and 
these two rarefactions reinforce each other. 

If you have followed the foregoing rather involved and 
detailed explanation, you can anticipate what happens next. 
By the time the pin has traveled on its return trip from A to 
C, the rarefaction has traveled to the bottom of the tube 



and has been reflected to the open end. An instant later the 
head of the rarefaction begins to emerge from the tube. This 
scarcity of molecules at the open end sucks in molecules from 
the still air outside the tube, and thereby the head of a com- 
pression starts down the tube. By the time the complete 
rarefaction has emerged, a complete compression has been 
formed, and this compression travels through the tube as the 
first compression did. Concurrently the pin rushes from C, 
past B, and comes to momentary rest at A, forming its own 
compression. Therefore the escaping rarefaction forms a 
compression, and the movement of the pin toward the mouth 
of the tube also forms a compression. These two compressions 
occur concurrently and reinforce each other. 

It is this working together of the vibrating source with the 
reflected wave which creates resonance in a wind instrument. 
This process has been described in detail because it is the 
cardinal principle of the wind instrument. It is easy to under- 
stand, from such a recital of what takes place, how the length 
of the tube and the vibrating source must work together. 
The tube must be of such a length that the reflected wave 
returns on time to meet the next impulse from the vibrating 
source. If the tube is too short or too long, the reflected wave 
returns too soon or too late to meet the next impulse coming 
from the vibrating source. When the reflected wave returns 
in phase with the vibrating source, it gives the vibrator a 
boost, because they are both pulling in the same direction. 
This boost, in turn, creates a stronger impulse in the vibrat- 
ing source, which is reflected with great energy to give a still 
bigger boost to the vibrator. 

This working together is something like two persons push- 
ing a stalled automobile. If one pushes and then the other, 

the car does not move far, if at all. But when both persons 
throw their weight against the car at the same instant, the 
combined force moves the car. 

When the pin moves from C to A, a compression is formed. 
When the pin moves from A to C, a rarefaction is formed. 
Since one wave consists of a compression and a rarefaction, 
the movement of the pin from C to A and back to C produces 
one complete wave. We have noted that during the time it 
takes the pin to travel forward and back, the compression 
travels twice the length of the tube (down and up), and also 
the rarefaction travels twice the length of the tube. One 
complete wave therefore travels four times the length of the 
tube during a single vibration. This is an important principle 
of a tube closed at one end. 

A tube open at both ends behaves like two closed tubes 
with their closed ends butted together. An impulse which 
enters an open tube does not travel the length of the tube 
and emerge out into the air from the opposite end. For some 
strange reason an impulse which enters an open tube always 
emerges from the same end by which it enters. Apparently 
the impulse entering one end does not have sufficient strength 
to emerge into the still air at the opposite end. The impulse 
travels down the tube, is reflected at the opposite end by the 
still air and starts back up the tube toward the end by which 
it entered. When the tube is of a proper length to give 
resonance to a certain rate of vibration, the wave is reflected 
up the tube where, at the very center, it meets a second wave 
coming from the top end. The two waves are repelled by each 
other at the center, and the second wave is reflected out the 
top end by which it entered. This meeting place in the center 
serves as a sort of booster station which gives the waves the 



needed energy to emerge into the still air at each end of the 
tube. The impulse entering the top end emerges from the top 
end, and the impulse coming from the bottom end emerges 
from the bottom end. 

In the closed tube at any one instant there is only one 
quarter of a wave length. At the open end is an antinode, 
where the molecules are in violent movement but where the 
pressure is slight. At the closed end is a node. Here the mole- 
cules cannot move, because the end is closed; but the pressure 
here is intense. 

In the open tube at any one instant there is only a half 
of a wave length. At the top end is an antinode, where the 





Fig. 5 

Drawing to represent difference between (A) closed pipe and 
(B) open pipe, sounding same fundamental note. Note closed pipe 
is just half as long as open pipe. Both pipes are open at top, and at 
open end of each is an antinode. The closed pipe is closed at bottom 
end, and here lies a node. The open pipe is open at both ends. In 
the center lies a node, and at the bottom end lies an antinode. 

molecules are in violent movement but where the pressure 
is slight. In the center is a node, where the movement of the 
molecules is at a minimum, but where the pressure is the 
greatest. At the bottom end is another antinode. 

This difference is sometimes represented diagrammatically, 
as in Fig. 5. 

In our illustration of vibration we used a pin because of its 
simplicity. In wind instruments the vibration is set up by the 
lips of the player, as in trumpet or trombone; by reeds, as in 
clarinet or oboe; or by eddy currents of air, as in flute or 

In all cases, the vibration itself is weak and not particu- 
larly musical. This can be demonstrated by blowing the 
mouthpieces alone, detached from the rest of the instrument. 
The tube comprising the rest of the instrument serves to 
build up the weak sound of the vibrating source, and not only 
to build it up and reinforce it until it sounds louder, but to 
refine the sound and make it more musical. 

Every wind instrument is made a certain length so it can 
build up and reinforce a certain note. When the lips of the 
player or other vibrating source produces this certain rate 
of vibration, the instrument builds it up, and the funda- 
mental or lowest note of the instrument is produced. Other 
rates of vibration may be produced by the lips of the player, 
but no sound is produced by these vibrations, or else they are 
very weak and distorted, because the reflected waves do not 
arrive on the exact instant to meet the new wave coming 
from the player's lips. The length of the instrument is such 
that only one wave is capable of making the round trip on time. 

This is what we mean when we say a cornet is built in Bb 
or a French horn is built in F or a flute is built in C. The 



instrument is built of such a length that it reinforces the wave 
length of this particular fundamental note. When the lips 
of the player are caused to vibrate against the mouthpiece of 
the horn, there may be other waves going out of the horn, 
because the lips do not have very accurate or exact control 
over these vibrations, but these other waves do not "fit" the 
horn. They may be reflected at the bell of the instrument so 
that they come back with their compression point coinciding 
with the point of rarefaction of the next wave. In this case, 
both are neutralized and no sound takes place. Or the points 
of compression and rarefaction may be just a little behind or 
a little in front of each other; in which case the tone is weak, 
stuffy and unpleasant. 

It is this even, periodical pulsation of certain frequency 
which produces a musical note. If notes of other frequencies 
were given resonance along with the note which "fits" the 
instrument, not music but noise would be the result. In the 
physics class, this point is often illustrated by a revolving 
disc pierced with two concentric rows of holes over which jets 
of air are placed. In one circle the holes are pierced equi- 
distant apart, and when the air jet is placed over this row of 
holes as the disc revolves, a clear musical note sounds. In the 
other circle the holes are pierced irregularly, with varying 
distances between holes. When the air jet is placed over this 
row of holes as the disc revolves, the result is an unpleasant 
noise. This jumble of vibrations of various frequencies pro- 
duces noise and not music. The same is true in a musical 
instrument. But the tube of a wind instrument picks out and 
amplifies only one wave whose length "fits" it, and kills off 
all the others. 

Here is a strange provision in nature which leads the 


poetical or philosophical mind to believe that sweet sound is 
"preferred" in nature. In a wind instrument there operates 
a physical law which stands guard over the sound and sorts 
out and obliterates those vibrations which are unmusical, 
amplifying only those which are pleasing. In the vibrating 
source, whether it be the lips of the player, the single reed, 
the double reed or the eddy currents of air, only those vi- 
brations are allowed to be amplified which are desirable for 
the pitch note. The tubing is a sort of resonance filter which 
lets only the proper vibrations pass. The others are dampened 
out. In this way a clear and beautiful note is produced out 
of the mixture of vibrations which originates at the vibrating 

There is a brief instant, a small part of a second, between 
the time when the vibration is started at the vibrating source 
and the time when the full, clear note is sounded. During 
this brief instant, the sorting or filtering process is going on. 
The mixture of vibrations starts down the tube of the instru- 
ment, is reflected at the bell and comes back to dampen or 
reinforce the vibrations which are being poured forth from 
the mouthpiece. Since there is only one wave which "fits' 1 
the length of tubing, the rate of vibration having this length 
is reinforced by the reflected wave and becomes louder. At 
the same time the waves which do not "fit" are undergoing 
a dampening-out process. They become weaker and weaker 
because when they are reflected at the bell they come back 
"out of step" with the waves which are issuing from the 
mouthpiece. The tendency, therefore, is to stop at the mouth- 
piece those vibrations whose waves do not "fit" the length 
of tubing in the instrument. Soon the vibration whose wave 
"fits" the length of tubing is dominating, and as it becomes 



louder, the other vibrations become weaker and are stopped. 
The wave which "fits" the tubing is pulsating violently, so 
much so that it reacts on the vibrating source and actually 
causes the lips, reeds or eddy currents to vibrate more 
vigorously than they did in the beginning. The wave even 
becomes master of the vibrating source and confines its 
vibrations to the rate of the wave. The note then speaks out 
clearly and in full voice. 

This filtering process takes place in a small part of a second. 
If the tubing is short and the rate of vibration is high, this 
interval is shorter than if the tubing is long and the rate of 
vibration is slow. For instance, in a two-foot flute sounding 
a fundamental of C = 261.6 vibrations per second, the inter- 
val would be short, for a rarefaction of a wave which does not 
"fit" doesn't have to meet a compression more than a few 
times until this wave is silenced. The reflected wave tries to 
come back out of phase once, twice or three times, and it 
becomes so weak that it dies. It not only does not sound, but 
it ceases to issue forth at the mouthpiece. At a vibration rate 
of 261.6 times per second, even ten attempts would require 
only that part of a second represented by dividing ten into 
261, or about ^ of a second. In long horns with low funda- 
mental notes of slow vibration, sometimes the interval is so 
long as to be noticeable. In defective instruments in which 
the resonating tube is not of exactly the proper proportions, 
sometimes there is heard a sort of crack or noise which indi- 
cates a struggle of the note to gain ascendancy over the 
others. Instead of a smooth sorting out of the various waves, 
there is a struggle between two or more, and finally the 
winning note cracks out. 

The diameter and taper of the bore of an instrument affect 


the length of the tone wave the tubing is capable of reinforc- 
ing. It has been found that for tubing two thirds cylindrical, 
with the lowest third of conical shape ending in a bell, a tube 
X inch in diameter must be 40^ inches long to sound At|, but 
that a similar tube only T\ of an inch in diameter must be 
41 1 inches long to sound the same note. One sixteenth of an 
inch in bore makes a difference of if inches in length. The 
trombone and the euphonium are built in the same pitch, 
but owing to the fact that the euphonium is of larger bore, 
it is not quite as long as the trombone. In a flute, with 
cylindrical tubing, this fact is not so pronounced. Here a 10 
per cent increase in the diameter of the tubing shortens the 
tubing only one per cent. 

The size of the bell is an important factor in determining 
the length of tubing which will produce a certain note. 
Acoustical engineers call it "end correction." It has been 
found that there is a certain cubic content of air outside the 
instrument which operates as an effective part of the column 
of air in the instrument. The amount of this air is determined 
by the diameter of the bell. In a cylindrical tube without a 
bell, this amount is about six tenths of the radius of the 
tubing. In other words, six tenths of the radius of the opening 
can be added to the actual length of the instrument tubing. 
For instance, if a tube is found to be 105 inches long and is 
4 inches in radius, the effective length of the tubing will be 
105 inches plus six tenths of 4 inches or 2.4 inches, giving 
a total length of 105 inches plus 2.4 inches, or 107.4 inches. 
In tapered tubing ending in a bell, the end correction is much 
more difficult to computate, but the principle still obtains. 

It should be said here that although the lowest note on the 
instrument is that which has a wave length twice its length 



(the clarinet excepted), it is not easy to play this low note in 
some instances. The horn may be capable, but the performer's 
lips may not be equal to the task. The upper octave of the fun- 
damental is more easily played, and the fundamental itself is 
less often used. In the case of the French horn, the first 
two octaves above the fundamental are difficult to play, and 
the usual playing range begins at the third octave. 

Resonance in a violin or other string instruments is differ- 
ent in principle from that in a wind instrument. Both types 
of instruments have this in common: the vibrating source is 
weak, and this weak sound requires reinforcing. The manner 
of obtaining this reinforcement is directly opposite in these 
types of instruments, however. In the wind instrument the 
resonating chamber, which is the tube, determines what note 
shall be given resonance and through the selecting process 
brings about this resonance. In the violin the note to be re- 
inforced is given to the violin by the length of string, and 
the sound chest merely reinforces this note. The sound chest 
has nothing to do with what note is given resonance: it 
only amplifies whatever note is given it by the string. An 
exception to this is the case of a "wolf" note, where the 
periodic vibration of the sound chest itself forces a note to 

In the string instrument it is easy to control the pitch of the 
notes. The string is pressed down against the fingerboard, and 
when its length is thus determined there is only one note 
which can result. The weak vibration of the string does not 
sound very loud, nor does it have much musical quality. 
The string cuts through the air without setting much of it 
into vibration. But when this weak impulse is transmitted 
down through the bridge to the belly or top of the violin, 


the whole top is set in vibration. When the top vibrates, it 
in turn sets into vibration the large cubic content of air in 
the sound chest. The back also vibrates in unison with the 
top, the impulse not only being transmitted through the air 
in the sound chest but through the sound post which touches 
both top and back. This greatly amplified sound, which 
originated weakly in the string, passes out through the / 
holes and is many times greater in volume than the weak 
vibration of the string. 

The main point to bear in mind in comparing these two 
types of instruments is that the sound chest of the violin is 
simply a magnifying agent, a sort of megaphone or sound- 
board. It takes the pitch of the string and amplifies it. The 
wind-instrument tube, on the other hand, actually de- 
termines the pitch of the note through the very process of 
amplification. In the wind instrument the source of the 
vibration is not easily controlled, hence the mechanism that 
reinforces the note must be under exact control In the string 
instrument it is the source of the vibration that is under 
exact control, and hence the mechanism which reinforces the 
note does not have to be under such exact control. That is 
why making the tube of the wind instrument is such an 
exacting job, while the violin body need not be made with 
such precision. 


We have been talking about one wave "fitting" an instru- 
ment. Perhaps you are wondering how the bugler manages 
to make several notes "fit" his bugle. He uses several notes 
to play his calls. If only one wave length will fit a given 
length of tubing, how can he play several notes on the same 



instrument ? That's a fair question, and we'll see if we can 
find the answer. 

Everyone knows that if a person takes hold of a rope at one 
end, with the other end tied to a tree, and shakes the end up 
and down, waves will form in the rope. When the rope is 
shaken very slowly, there will be but one long, complete 
wave. If shaken twice as fast, two waves will form in the 
rope. The faster the rope is shaken, the more waves there will 
be in the length of the rope. 

Fig. 6 

Showing how the tone wave divides and produces the open tones 
of the natural harmonic scale. 

This illustrates what happens in a wind instrument. When 
the lowest note is sounded, one complete wave twice the 
length of the instrument results, which is C on the instru- 
ment. When the lips are tightened and the instrument is 

blown harder, the vibration rate is increased until the one 
complete wave divides up into two waves, each just half as 
long. The resulting note is an octave higher than the former. 
By holding the lips tighter and blowing still harder, three 
waves result, each one third as long as the first, and the note 
produced is G. When the lips are tightened still more and the 
instrument is blown still harder, there are four waves, each 
one fourth as long as the first, and the note sounded is C. 
When five waves are produced, E is sounded, and when six 
are produced, G above the staff is sounded. These are the 
notes the bugler has at his command, but he doesn't use the 

If the performer were equal to the task he could go on and 
on, compelling the wave to divide into seven, eight and up to 
twenty or more separate waves. (See Fig. 7.) These notes 
would all be the harmonics of the fundamental, and the 
resulting scale would be the natural, or harmonic, scale. The 
notes in this harmonic scale are called partials, the lowest 
being the first partial, the octave being the second partial, 
the next one (G) being the third partial, and so on. These 
notes are also spoken of as the "open" notes, no slides, 
valves or keys being necessary to play them. It is easy to 
see (Fig. 7) that there are wide gaps in the lower parts of 
the scale, but that as the higher partials are reached the 
notes are close together until, when the thirteenth partial is 
reached, the notes are a half-step apart. In this upper range 
we have all the notes we need, but in the lower range there 
are some wide gaps that must be bridged over in some way. 

The lowest note any tube can give resonance to has the 
lowest frequency for that tube. If we gradually increase the 
frequency, the note will fade out until the tube has no 









J-< CO ,1 -rt 
q,) QJ < L ^ 

SS'rj 5 3 

s c^ s 


c ^ 




S? o' 

c 9 o x 
P fe ^o 

3^ S 


resonance. If the frequency is gradually increased until it is 
doubled, a rate of vibration will be reached which will be 
built up by an open tube. What happens is this: The waves 
cannot quite make the round trip on time as the frequency 
is increased. Their efforts, as they are reflected, become more 
and more futile because they become farther and farther 
"out of step." Finally resonance stops. Then, as the fre- 
quency increases still further, the waves are able to set up 
substations, as it were. Although they cannot make the 
complete round trip on time, they shorten their travel, 
cutting the distance in half. Two waves, then, each half as 
long as formerly, span the length of the tube. These two 
halves, each traveling concurrently, make the shorter trip on 
time. The pulsations do not travel faster their speed is 
constant but they cut the trip in half. There are two waves 
now, each half as long as at first. The frequency is twice what 
it was formerly, the waves are just half as long, and the pitch 
is raised an octave. 

This is true of open pipes only. On the closed pipe the first 
note above the fundamental is reached when the frequency 
is three times that of the fundamental. Instead of one wave 
there are now three, and each wave is one third as long as 
that of the fundamental. The note sounded is a twelfth above 
the fundamental. This third partial is sounded, both on the 
open as well as on the closed pipe, although of course the 
open pipe is twice the length of the closed pipe for this note. 

When the frequency becomes four times that of the funda- 
mental, the length of the open pipe is spanned by four waves, 
each one fourth as long as that of the fundamental, and the 
note sounded is two octaves above the fundamental. This 
even partial cannot be reproduced on the closed pipe. The 









6 G 

,3 G 



Diagram showing how waves break up into smaller waves to form 
various notes of the harmonic scale. Observe that the length of the 
wave for the second partial is half that of the first, or fundamental; 
that the length of the wave of the third is one third that of the 
fundamental; that the length of the wave of the fourth is one 
fourth that of the fundamental; etc. Also observe that the even- 
numbered partials are missing on the closed pipe. 

next partial for the closed pipe is reached when the frequency 
is five times that of the fundamental. The wave is one fifth 
that of the fundamental, and the note sounded is two octaves 
and a third above the fundamental. This fifth partial is of 


course produced on the open pipe also, although the pipe is 
twice as long as the closed pipe for this note. 

It will be seen, therefore, that both even and odd partials 
are produced on the open pipe, whereas only the odd partials 
the third, fifth, seventh, etc., can be produced on the 
closed pipe. 


The bugle is limited to the harmonic scale of open notes 
and can execute its calls because the music is written es-, 
pecially for it; but to play notes other than these open notes, 
we must resort to some means to get the notes in between. 

In the manufacture of wind instruments this is done in 
various ways. In the trombone, slides are provided which 
lengthen the tubing enough to produce the notes to fill in. 
When the slides are closed, or are in " first " position, the 
trombone can sound the open notes. By extending the slides 
about four inches to the second position, enough tubing is 
added to make the next half-tone below this open note. By 
extending the slides another position, more tubing is added 
and another half-step is added, making the note a full tone 
below the open tone. By using all seven positions, it is 
theoretically possible to play six steps below the open notes. 

The author is well aware that the trombone is not treated 
in music as a transposing instrument, but the method of 
treatment here has been chosen to eliminate confusion in 
comparing it with the harmonic scale of open notes used as 
illustration for all instruments (Fig. 7). 

From the fundamental, or lowest open note, to the second 
partial, which is the octave of the fundamental, there is a 
gap of twelve half-tones which are partially bridged by the 
use of the slides. The second partial is sounded and sufficient 



length of tubing is added by extending the slides to secure a 
total length of tubing capable of producing the desired notes. 
However, the seven positions of the slides are able to take 
us down only six half-steps, or down to F$ (sounds Et)), 
leaving a gap of notes between F$ and the lower C which 
cannot be produced. Composers who write music for the 
trombone bear this in mind and do not call upon it to play 
these notes, giving them to other instruments that can. 

The bass trombone is called upon to play these notes, but 
it has an extra length of tubing which can be added by 
operating a valve with the left thumb. This lowers the pitch 
of the trombone five, and sometimes six, half-tones. By 
lowering the pitch of the trombone five half-tones the six 
slide positions then can be added, making it possible to play 
down to Db (sounds B). Then the thumb valve is closed by 
releasing the lever, and the open C (sounds Bb) can be 
played, completing the chromatic range of twelve half-tones. 

In the second octave above the fundamental we have one 
open note which we can play without the use of the slides. 
This note is G and is the third partial in the natural scale. 
It is produced when the fundamental wave breaks up into 
three parts. This makes the following open notes available: 
C, G and the octave C, all of which are notes of the natural 
harmonic scale. To play Qf, D, D#, E and F, which lie 
between the open tones C and G, it is necessary to use the 
slides, descending from the G and adding lengths of tubing 
necessary to make each corresponding and lower note. 
Between G and C above, the G#, A, A# and B are lacking 
from the natural scale and must be played by adding neces- 
sary lengths of tubing to the upper C, or fourth partial. 

In the third octave above the fundamental we have more 


notes that lie in the natural scale, and it is not necessary to 
make so many of them by using the slides. In this octave we 
have E, G and Bb (quite flat), besides the upper and lower 
Cs. The fourth octave above the fundamental is out of the 
range of the trombone, but we know how these notes would 
have to be played if it were possible for the player to produce 
them in this octave. Here we have a more complete natural 
scale comprised of C, D, E, F#, G, A, Bb, B and C. In this 
octave the slides would need to be used to make only some 
half-steps. To produce C#, for instance, it would be necessary 
to play the open D of the natural scale and drop the slides 
to the second position, thereby adding a length of tubing 
necessary to produce the lower note. 

As stated before, the scale becomes almost complete as 
the upper part of the harmonic scale is reached, but it is im- 
possible to produce these notes on some instruments. 

The same natural scale as on the trombone exists on the 
cornet or trumpet or any other wind instrument, except the 
clarinet, but the C of each scale must be the C of the instru- 
ment, depending upon the key in which it is built. If the 
instrument is built in F, C of the natural scale for this 
instrument will sound F, If built in Bb, C will sound Bb; 
and so on. The relation between the notes of the natural scale 
is the same, however. 


On valve instruments the valves take the place of the 
slides on the trombone. The gaps of the natural scale are 
bridged by adding various lengths of tubing to the open 
notes. The first valve on the cornet or trumpet opens up a 
length of tubing which enables the performer to play a whole 



tone lower than the open note. The second valve opens up a 
half-tone, and the third opens up three half-tones or a tone 
and a half. 

By combining the various valves, more than one and a 
half tones may be added. By combining the second and third 
valves, a length of tubing is added equal to two full tones; 
by combining the first and third, two and a half tones are 
added; by combining all three, three tones are added. These 
combined lengths of tubing enable the performer to go six 
half-tones below any open note and are sufficient to bridge 
all the gaps in the natural scale except in the first octave 
above the fundamental, in which only six of the twelve 
half-tones are available. 

These six half-tones are sufficient in smaller-bore instru- 
ments such as cornets, trumpets, alto horns and baritones, 
because the fundamental is seldom used. The lowest partial 
called for in music for these instruments is the second, or C, 
an octave above the fundamental. With the valves it is 
possible to play six half-tones below this second partial or 
down to F#. But the fundamental is playable on larger-bore 
instruments such as the euphonium, bass tuba and sousa- 
phone. Therefore, on some of these a fourth valve is added 
which controls an extra length of tubing capable of lowering 
any open tone five or six half-steps. The four valves have a 
combined range of twelve half-steps, which are sufficient to 
lower the second open note a full octave, thereby bridging 
the gap chromatically from the fundamental upward. 


On flutes, oboes, bassoons and saxophones, the same gaps 
in the scale are found, but these gaps are bridged in a different 


manner by a system of keys. When all holes are closed and 
the lowest note is sounded, one wave lies in the instrument, 
one half of the wave extending from the mouthpiece to the 
bell, the other half being a reflected wave extending from 
the bell to the mouthpiece. As each lowest hole is opened, the 
wave is successively shortened. In this way it is possible to 
play chromatically from the lowest note a full octave up- 
ward. Then the instrument is "overblown," changing from 
the fundamental, or first partial, to the second partial, an 
octave above. To assist the player to obtain this second 
partial, an octave-key hole is opened. With all keys closed 
and with the octave key open, it is possible to sound this 
second partial. Then, successively opening each lowest hole, 
it is possible to play chromatically up the second octave. 
The same fingering is used in the second octave as in the 
first, the fingered notes sounding an octave above because 
they are based on the second partial This is theoretically 
the principle of all such "open pipe" keyed instruments. 
Actually there is a slight deviation from this principle in 
the second octave. In the saxophone, for example, from D 
to G# in the second octave is played with the first octave 
key open, but beginning with A, a second register key higher 
up is opened. 


The clarinet is peculiar in that its natural scale comprises 
only the odd partials. The length of its fundamental tone 
wave breaks up into three parts but will not break up into 
two or four; it breaks up into five parts but will not break 
up into six. That leaves gaps in the natural scale which are 
not found on other instruments that have both the even and 



odd partials. Between its fundamental note and the first 
occurring partial is a gap of nine and a half whole tones, or a 
twelfth, which must be bridged. This is the third partial in 
the natural scale, the second, or octave, being absent. 

The second, fourth, sixth all the even partials are missing 
from the natural scale of the clarinet, leaving wider gaps 
than on all other wind instruments that must be bridged. 
Since there are wide gaps in the natural scale of the clarinet, 
a great number of keys are necessary. 

As in the saxophone or flute, when all keys are closed, 
the lowest note, or first partial, is sounded. Then, successively 
opening each lowest key, the player can ascend chromati- 
cally. Since the second partial is missing on the clarinet, it is 
necessary to bridge the gap from the first to the third partial, 
a gap of a twelfth. Then a register key is opened to assist the 
player in sounding the third partial. The ascending notes from 
here are then based on the third partial, and each lowest 
hole can then be opened in ascending chromatically upward. 


There are apple pies and apple pies. People go through 
life eating apple pies, and they are never a bit uncertain 
that each apple pie is an apple pie. There is a difference, 
however, among them. Some are good, some are fair, and 
some are bad. What makes the difference is the quality and 
proportion of the ingredients. Some may lack butter, some 
nutmeg, some sugar. But they are all apple pies. 

That is just the way with some musical instruments. They 
may blow a certain note in pitch, but the quality is gone. 
The note is dead, it lacks richness, beauty, flavor. The 
skeleton of the tone is there, but the sweetness is gone. 



We shall now try to find out what it is that accounts for 
this tone quality in an instrument. 

Do you know why telephone operators are instructed to 
trill their r when repeating the number three? It may seem 
strange, but the reason for this lies at the basis of tone 
quality in horns. 

It has been found that "two" and "three" often sound 
the same over the telephone. The vowel sound in "two" is 
like the vowel sound in "moo," and the vowel sound in 
"three" is like the vowel sound in "bee." These two vowel 
sounds do not ordinarily seem similar, but the telephone 
mechanism sometimes dampens out the distinguishing 
characteristic of the vowel in "bee" until it cannot be dis- 
tinguished from the vowel sound in "moo." 

It has been determined that the common characteristic 
is a group of overtones which lie on either side of 300 vibra- 
tions per second, but the vowel in "bee" has an additional 
group which lies near 3,000 vibrations. This high group is 
lacking in the vowel in "moo." 

In other words, take the vowel in "moo" and add the 
high overtones and the result is the vowel in "bee." The 
telephone mechanism is sometimes incapable of reproducing 
the high overtones, and consequently the vowel sounds as 
in "moo" instead of as in "bee." 

It may seem strange to think of vowel sounds as being 
composed of combinations of notes of various rates of vibra- 
tion, but scientists have been able to produce all the vowel 
sounds accurately by machines capable of sounding variously 
pitched notes with the various intensities at the same time. 
By using these machines it is possible to set various vibrators 
going together, and by selecting the proper vibrators and 



by regulating the intensity of each these machines can be 
made to "talk/' 

It is a common observation that it is difficult, if not im- 
possible, to sing distinctly such vowel sounds as in "moo" 
and in "mow" at a very high pitch. It is equally difficult 
to sing distinctly such vowel sounds as in "bee" and in 





Fig. 9 

Illustrating that the characteristic in the vowel sound in " three " 
is often filtered out when transmitted over the telephone and 
comes out at the receiver sounding like "two." This characteristic 
that is lost has been found to be a high overtone whose vibration 
rate is about 3,000. The rest of the vowel sound in " three " is the 
vowel sound "two," which is a group of overtones around 300 
vibrations a second. It is such overtones as these which regulate 
tone quality in the horn. 


"bate" at low pitch. To get the characteristic overtones of 
the former, the pitch at which they are uttered must be 
low, so that the low overtones will be produced. To get the 
characteristic overtones of the latter, the pitch must be high, 
so that the high overtones will be produced. 

This is the basis of all the discussion about "opera in 
English." When the opera is translated from the foreign 
tongue into the English, the vowel sounds which the music 
indicates must be sung at high pitch may be such that their 
distinguishing characteristic is a low overtone. On the other 
hand, vowel sounds that must be sung at low pitch may 
happen, through translation, to be those whose distinguishing 
characteristic is a high overtone. It is physically impossible 
to sing such notes and utter the vowel distinctly at the 
same time. 

What has this to do with quality of tone in musical in- 
struments? Just this: It is the presence or absence of these 
overtones that determines what the quality of tone of the 
instrument shall be. Just as the presence or absence of the 
high overtone makes the difference between the vowel sounds 
in "moo" and in "bee," so the presence or absence of the 
proper overtones in the instrument makes the difference 
between a sweet, rich tone and a flat, ugly tone. 

The presence of overtones can be graphically illustrated 
on a stretched string, such as a violin string. When bowed 
in a certain place and manner the string can be made to 
vibrate as a whole. This tone is dull; such a tone is that 
produced by the beginner. The more skillful player can make 
the string break up into smaller sections while vibrating as 
a whole. The pitch is that of the longest tone wave because 
it predominates over all the others, but the other smaller 


sections are vibrating as units and producing overtones of 
less energy. These overtones blend in with the fundamental 
tone, producing a richness of tone quality. The more of these 
overtones that can be produced in the proper relation to the 
fundamental, the richer the tone quality is. 

The same is true of the wind instrument. Some players 
haven't the trained lip control, or embouchure, which en- 
ables them to play a note so that the tone wave vibrates 
as a whole and in many smaller sections at the same time. 
Often the inferior instrument is defective, certain spots in it 
serving to kill the overtones where they should appear. It is 
very important that the instrument be built to accommodate 
these overtones, as their presence determines the tone 

Although the baritone and euphonium are approximately 
the same length as the trombone, difference in bore makes a 
difference in tone quality. The euphonium is the broadest 
and fullest in tone, the baritone is brighter in tone, and the 
trombone is the most brilliant of the three. The difference in 
bore regulates what overtones are possible on the instrument, 
and we recognize this fact by listening to the characteristics 
of each. Each instrument may possess the same number of 
overtones, but in the large-bore horns it is the low overtones 
that are produced while the high overtones are dropped out, 
and in the small-bore instruments it is the high overtones 
that are favored. This results in difference of tone quality 
of the euphonium, baritone and trombone. 

Not only the width of bore influences the tone quality, 
but the taper in the bore is equally influential. In general, 
the wider or more tapered the bore, the fewer overtones 
there are. Conversely, the more nearly cylindrical the bore, 


the more overtones there will be. But while we say it is the 
width of bore and the taper which affect the tone, it would 
be more accurate to say that the bore and taper regulate 
the production of overtones, which determine the tone 

A beautiful illustration of the fact that overtones govern 
tone quality in instruments is given in comparing the mello- 
phone and the French horn. The French horn is about twice 
the length of the mellophone and should play an octave lower. 
As a matter of fact, the playing range is about the same. 
The tone quality, however, is different, the French horn 
having an unusually beautiful and rich tone. The fundamen- 
tal tone of the French horn is an octave below that of the 
mellophone, but the French horn player doesn't use the 
lower range of his instrument, confining his playing to the 
upper range. The playing range on the mellophone being 
the same as that on the French horn, and its fundamental 
or lowest note being an octave higher, it is evident that the 
mellophone uses a lower part of its range than does the 
French horn. 

Now we have seen that, as we ascend from the fundamen- 
tal note of a horn, the overtones become closer together until 
in the upper ranges the scale is almost complete. Since the 
French horn uses its upper range to play the same notes the 
mellophone plays in its lower range, we can readily see that 
there are many more overtones present in the French horn's 
tone than in that of the mellophone. 

That overtones account for quality of tone has been 
demonstrated in various ways. A German scientist by the 
name of Helmholtz invented a rather complete set of resona- 
tors which looked like a lot of Christmas-tree ornaments. 



They were so made that each would vibrate in sympathy 
with a certain pitched tone, shutting out all others. When 
one end was inserted in the ear, the resonator would pick 
out its particular note from a complex and rich tone and 
would vibrate violently, causing its note to sound loudly 
in the ear. By the use of these resonators, rich tones were 
analyzed into the fundamental and all the overtones that 
were present. 

Now if you are good at fractions you can look at this sub- 
ject of overtones in a little different way. If fractions and 
ratios and other arithmetical calculations bother you, just 
skip the next few paragraphs. But after all, music and har- 
mony are all tied up in such considerations as fractions and 
ratios and can hardly be understood without a little figuring. 
A little effort expended in wading through the following few 
paragraphs will pay handsomely in the understanding of this 
fascinating subject of the harmonic scale and the overtones. 

Quality of tone is not due entirely to the number of over- 
tones: they must be the overtones of the harmonic scale. 
When an overtone occurs which is not in the harmonic scale 
of the fundamental, an inharmonious sound results. What we 
call "noise" is a combination of notes of various rates of 
vibration that are in an inharmonious relation to each other. 
That means that the lowest note in the combination is 
combined with other notes whose wave lengths are NOT %> 
}i> %> Y^ etc -> f * ts wave length. These other notes may 
be ^ or ^ of the wave length of the lowest note. The result 
of sounding such a group is a harsh, unpleasant noise, 
especially if sounded loudly. 

A professor of physics once used an ingenious illustration 
to show this harmonious relation of the partials of a musical 


instrument. He fashioned some blocks of wood, each so cut 
to size that it had a definite pitch. He dropped the blocks 
successively on a table and thereby played an ascending 
harmonic scale of C, G, C, E, G, C. Then he showed that 
when he dropped all the blocks together the sound was a 
sort of musical chord. He also had other blocks whose pitch 
stood in odd ratios to each other, not of X, >, %, J-f, etc., 
but of yz>y< y*]y etc. When these were dropped together 
on the table, the result was not a musical sound, but only a 
dissonant sound which we call noise. 

This same principle can be illustrated on a revolving disc 
in which concentric circles of holes are bored and which can 
be "played" by air jets, one jet to each circle of holes. Let 
the circle nearest the center have 16 holes, the next ones 18, 
20, 22, 24, 27, 32. If the circles having 16, 20, 24 and 32 
holes be played together, a pleasing chord is sounded : 

A little arithmetic will readily show these numbers to stand 
in relation to 4 as 4, 5, 6 and 8 respectively. These notes 
therefore are the fourth, fifth, sixth and eighth partials of 4, 
which can be considered as the fundamental. If the disc 
revolves 10 times per second, the fundamental would have a 
frequency of 40 and the partials up to the eighth would have 
a frequency of 40, 80, 120, 160, 200, 240 and 320 respectively. 
Now, taking 40 vibrations as a fundamental, all of these 
partials are related as I (fundamental), 2, 3, 4, 5, 6 and 8. 
Also, taking the note with a frequency of 40 as the funda- 
mental wave, the wave lengths of these partials are related 


as I (fundamental), fa fa X, fa y 6 and #, For our pur- 
pose we will deal only with the fourth, fifth, sixth and 
eighth. When these are sounded together a pleasant chord 
results, as related above. But if the other circles of holes are 
sounded together, that is, those circles having 18, 22 and 27 
holes, only noise is produced; and if they are sounded with 
the others, the pleasant chord is turned into noise. Applying 
our simple arithmetic, we readily see that the circles having 
1 8, 22 and 27 holes do not stand in simple relation to 4 as 
K> %> *A> Y< and so on, but as &, & and T&. These odd 
fractions of 4 produce noise and not harmony. 

Since the tube of a wind instrument, as shown previously, 
gives resonance only to those partials which are fa fa fa 
etc., of the fundamental, these odd and irregular partials, 
such as A, ^ and -fa in our illustration, cannot sound, and 
therefore only the musical overtones can occur. Here is an- 
other amazing provision of nature which gives us harmony 
in wind instruments and dampens out all inharmonious 
overtones which would produce noise. 

Some marvelous machines have been invented which reveal 
in an even more interesting way the presence of overtones. 
These machines are called oscillographs, and they picture 
sound by means of lines which look something like a sales 
curve or the fluctuations of the stock market when drawn 
on a graphing chart. For instance, the note of the tuning 
fork for A2-no may look like the top curve marked A in 
Fig. 10, while the note from the tuning fork for Ai-22O, an 
octave higher, would look like the curve marked B. The note 
from the tuning fork for -330 would look like the curve 
marked C. 

You will observe that the first curve has but one "hump," 


while the curve marked B has two complete "humps." This 
you would expect, since the second note is an octave higher 
than the first and has twice as many vibrations per second. 
The distance from left to right in our illustration represents 

Fig. 10 

a certain interval of time, in this instance y-Jv of a second, 
since only one "hump," representing a complete vibration, 
is shown for a note which has no vibrations per second. 
To show what the oscillograph records in a second would 
require a drawing no times as long as the illustration in 
Fig. 10, and there would be no "humps." Since the second 
note is an octave above the first and therefore has twice the 
frequency, in a second the oscillograph would show 220 
"humps." Since our illustration records what happens in 
rhr of a second, there are two "humps" in our illustration. 
The third curve, marked C, has three "humps" and there- 
fore is a recording of a note which has three times the fre- 
quency of the first note, or a twelfth above the first note. 

With this elementary understanding of how the oscillo- 
graph functions, we can now look at some more complicated 



curves as they are recorded when notes are sounded which 
have overtones in them. The three curves in Fig. 10 are of 
pure notes without overtones, such as come from the tuning 



Fig. II 

fork. Such notes are soft and definite to the ear, but are 
rather colorless. It is the presence of overtones which gives 
color, or timbre, or quality to notes. 

In Fig. ii we show the recordings of AJ-IIO and A r 22O, 



first separately and then when combined. When the two 
notes are blended together by being sounded simultaneously, 
the oscillograph curve looks like the curve shown marked 
A + B. Below this we show the curve for the pure note 
-330, and below this the curve which records the sound 
when all three notes are sounded together. This curve is 

Fig. 12 

marked A + B + C. This curve begins to look like the notes 
from musical instruments which have several overtones 
sounded together. This synthesizing of the pure tones from 
the tuning fork will help us understand the meaning of the 
curves for notes from musical instruments. 

In the top curve in Fig. 12 we see the oscillograph of a 
note from the flute when played softly. Here only the funda- 



mental of the note sounds, the overtones or harmonics not 
being emitted, at least not strongly enough to be recorded 
by the oscillograph . Such a curve is much like that of the 
tuning fork, and the soft flute note sounds to the ear much 
like the note from a tuning fork. But when the flute is blown 
loudly, the curve changes character, as seen in the bottom 
curve, because the overtones are strengthened. Their pres- 
ence is shown by the smaller humps in each of the larger 
humps. That the same pitched note is sounded in both cases 
is shown by the fact that in the given length in our illustra- 
tion there are three humps in each curve, proving that each 
note has the same frequency. The smaller humps in each of 
the larger humps indicate the presence of the harmonics 
blended in to give color, or timbre. Such a note seems 
brighter to the ear than the soft flute note. 

Fig. 13 

In Fig. 13 three cycles of a note from the oboe are shown 
as recorded by the oscillograph. Since this illustration is 
drawn on the same scale as that in Fig. 12, the three cycles 
in each show that both instruments are sounding the same 
pitch note. The curves differ only in characteristic shape. 
Observe that the oboe curve is considerably more compli- 
cated than the curve of the flute note sounded loudly. This 
intricacy of pattern indicates the presence of many har- 

monies. Scientists who have analyzed oboe notes find they 
may have six, eight, or even more harmonics. Usually the 
fourth and fifth harmonics are stronger than the funda- 
mental, with the sixth harmonic next in strength. Any two 
notes may have the same harmonics present, but their rela- 
tive strengths, if different, will make the notes sound different 
in timbre, or quality. 

Fig. 14 

In Fig. 14 three cycles of a note from a clarinet are shown 
as recorded by the oscillograph. Since the graph in Fig. 14 
is on the same scale as the graphs in Fig. 12 and in Fig. 13, 
these three notes are of the same pitch. Their only difference 
is in the number and strength of their harmonics, which 
account for their characteristic difference in tone color. 
Acousticians would also add another factor their phase 
but we'll skip that, for it makes the explanation much more 
complicated and much less easy to understand. Observe the 
erratic ups and downs of this curve pattern, but notice that 
each cycle is practically the same in shape. Analysis of the 
clarinet tone reveals that the first and third harmonics are 
usually prominent and of about equal strengths. The eighth, 
ninth and tenth partials are also fairly strong. Other har- 
monics may be present besides these, but they are weaker 
and sometimes do not show up. 



A note from the French horn records as shown in the curve 
in Fig. 15. Three cycles are shown, and since this illustration 
is on the same scale as those for the flute, oboe and clarinet, 
the note sounded is of the same pitch. The curve, however, 
differs in character from that of the flute, oboe or clarinet. 
This difference is of course due to the number and strength 
of the harmonics reproduced by the French horn, which give 
it its characteristic tonal coloring. When the tone is analyzed, 

Fig. 15 

the fundamental, the second and the third harmonics are 
usually prominent, their strength varying in descending 
order, from I to 3. But these are not the only harmonics 
present. In fact, it is the presence of many harmonics which 
makes the tone of the French horn so peculiar in tonal 

Since the curve of the note from the bass tuba, shown in 
Fig. 16, is drawn on a different scale from the curves of notes 
from the flute, oboe, clarinet and French horn, we cannot 
make a comparison as to pitch but can only notice its charac- 
teristic shape. This curve is comparatively simple, showing 
that it is made up of but few harmonics. Only one hump, or 
cycle, is shown, the smaller humps indicating the harmonics. 

As illustrated in the flute curves, the loudness with which 
an instrument is sounded has something to do with the 
harmonic mixture for any note. The curves in these oscillo- 


graph records are shown as being fairly characteristic of each 
instrument. Loudness may change any of them in varying 
degree. The notes in different parts of the scale on any one 
instrument differ in pattern also. The ear detects this differ- 
ence in some instruments more than in others, and the curves 

Fig. 16 

in different registers vary more in some instruments than 
in others. The middle register of the clarinet, for instance, 
is of different tonal coloring than the top register. But while 
there are these variations, each instrument has its character- 
istic curve pattern. Two Irish setters may not look exactly 
alike, but nobody would have any difficulty in telling them 
from two collie shepherd dogs. There is also a general family 
resemblance in curves from different instruments. Some of 



them are very similar to others, but experts who study these 
characteristic curve patterns can tell them apart. 

Looking at the question of tonal quality from the opposite 
angle, we would hardly expect the same sound to come from 
a bassoon and a French horn, but sometimes it is difficult 
to distinguish them. If these notes are recorded on an 
oscillograph, we find a definite similarity in the two curves, 
for it is the presence in both instruments of the same har- 
monics in the same relative strengths which makes this 
similarity in tone quality. When the French horn is blown 
vigorously it often sounds much like a trombone. This blast- 
ing of the French horn accentuates certain harmonics, which 
makes the oscillograph curve look something like that of the 

Tone quality is simply the result of mixing different har- 
monics in different proportions. Some instruments are built 
so they reproduce or give resonance to certain harmonics 
rather than to others and give greater power to some of these 
than to others. Just as a painter selects different pigments in 
varying proportions and mixes them together, producing a 
distinctive blend of coloring, so the different instruments 
select the various harmonics and temper them in degree of 
loudness, producing a distinctive tonal quality. It is this 
difference in the selective process which makes each instru- 
ment sound different from another. 


An apple bud grafted on a peach tree produces apples and 
not peaches. 

Musicians should take this tip from the horticulturist in 
choosing their mouthpieces, because the mouthpiece stands 


at the head of the instrument and is an important factor in 
determining how the instrument performs. 

This point is strikingly illustrated in the following ex- 
periment. The embouchure was removed from a flute, and 
an alto-clarinet mouthpiece was substituted. The flute 
immediately took on all the characteristics of the clarinet. 
It was no longer a flute. It was capable of producing only 
the odd partials of the natural scale, possessed the same 
gaps in the natural scale, no longer played in octaves as does 
the flute, and acted up in general like an incomplete clarinet. 

This is just like grafting a clarinet bud on a flute tree! The 
result is clarinet fruit hanging from a flute limb. 

This illustration brings up one of the most interesting 
questions in wind instruments, a question which is now 
being heatedly debated and which is a long way from being 
satisfactorily answered. Why does the addition of a single- 
beating reed to a cylindrical pipe make the resulting instru- 
ment perform like a closed pipe? The body of the flute is 
practically cylindrical. With the flute embouchure, or mouth- 
piece, the instrument behaves like an open pipe. When a 
single-beating reed mouthpiece is substituted, the instrument 
changes to a closed pipe. Does the reed close the end, making 
it a closed pipe such as a closed organ pipe ? The reed, how- 
ever, is open half the time and closed half the time. 

As a matter of fact, although the wave of the fundamental 
is four times the length of the instrument, and although 
the second and fourth partials cannot be utilized in the scale, 
the instrument still preserves some of the characteristics of 
the open pipe. 

The clarinet is substantially like our hybrid instrument 
described above, having a cylindrical body and a single- 



beating reed. Delicate analyses of the clarinet tone reveal 
that the second and fourth partials are present, although 
they are usually very weak and cannot be used in the scale. 
The eighth and tenth partials are quite strong as strong, in 
fact, as the ninth. These, however, are so high in the har- 
monic scale that they cannot be used. 

That the single-beating reed is not the only factor is 
shown by the fact that a cup mouthpiece actuated by the 
buzz of the lips also makes a cylindrical tube behave like a 
closed pipe. The slides of a trombone are cylindrical, and if 
a cup mouthpiece be fitted to them the resulting instrument 
behaves like a closed pipe. Analysis of the tone of the instru- 
ment, however, reveals the presence of the even partials, 
although they are very weak, as on the clarinet. The buzzing 
lips also close the pipe, just as does the reed, although the 
opening is open half the time and closed half the time. The 
blowhole of the flute, however, is open always, and the 
cylindrical tube alone, on the flute, is not sufficient to pro- 
duce a closed pipe. The closed pipe, therefore, seems to 
require a coupled system of cylindrical tubing and a device 
for closing the end. The fact that the reed on the saxophone 
and the cup mouthpiece on the cornet do not produce a 
closed pipe can be laid to the fact that in these instruments 
the cylindrical tubing is lacking, the tubing of both being 
conical in general shape. 

The layman has an idea that wind instruments are 
"blown." The usual practice of the layman in picking up a 
wind instrument is to blow a blast through the mouthpiece. No 
sound results, for such instruments are not "blown." The sound 
is produced by creating a vibration in the mouthpiece, this vi- 
bration being given resonance by the tube of the instrument. 


The flute comes nearest to being " blown" of all wind 
instruments, but this blowing is of a distinct kind. A thin 
stream of air is directed by the lips against the edge of the 
blowhole. How this produces a vibration is illustrated by a 
flexible stick in a stream of water. Everyone has seen such a 
stick vibrate or has felt the vibration when the stick is held 
in the hand. The water pushes the stick to one side. When 
displaced to one side the water rushes past, and the pressure 
against the stick on this side is relieved. But when this 
occurs, the pressure becomes great on the other side, and the 
stick is pushed back. So the stick is made to oscillate back 
and forth with regular periodicity. In the flute, the edge of 
the blowhole is stationary. The air rushes by the edge, out- 
side the flute. This rush of air raises the pressure outside 
and lessens the pressure inside. The stream of air, therefore, 
changes its course to the low-pressure area of the inside and 
is deflected inside. But, again, pressure is built up inside 
while the outside becomes rarefied. The air then reverses its 
course and escapes outside the flute. In this way pulsations 
of air are sent down the flute and a vibration is set up. 

In the cup-mouthpiece instruments the lips are stretched 
across the cup and intermittent puffs of air are permitted to 
escape between the opening of the lips. These pulsations of 
air set up a vibration in the instrument. The same sort of 
pulsations are created in reed instruments. In the single-reed 
instruments, intermittent puffs of air are permitted to pass 
into the instrument as the reed opens and shuts on the table 
of the mouthpiece. In double-reed instruments, the same sort 
of vibration is created by producing puffs of air as the two 
reeds open and close against each other. 


Afranio, 95, 102, 234 
Albert, clarinet system, 124 
Allen & Thorn, 274 
Almenrader, 102 
Amati, 41 

Armstrong, Louis, 142 
Arne, 120, 122 
Athenaeus, 275 

Babcock, Alpheus, 274 

Bach, Emanuel, 277, 284 

Bach, J. C., 120, 122, 286 

Bach, J. S., 28, 51, 91, 98, 103, 155, 

162, 164, 190, 196, 213, 277 
Balthasarini, 23, 48, 64, 97, 213, 290 
Band, composition of, n 

instrumentation of, 14 

seating arrangement of, 13, 182 
Band Instruments, American, 250 
Banjo, 295 
Barrett, 101 
Bass, figured, 22, 281 
Basset horn, 30, 127 
Bassoon, 86, 95, 102, 108, 234 

contra, 32, 96, 103, 218 

quinte, 92 
Baton, use of, 282 
Beethoven, 30, 99, 106, 122, 169, 197, 

217, 236, 256, 282 
Beiderbecke, 138 
Bellini, 220 

1 359} 

Bells, 263 
Bellstedt, 155, 183 
Bergonzi, 41 
Berigan, Bunny, 142 
Berlioz, 33, 54, 95, 100, 107, 128, 143, 
198, 219, 221, 238, 257, 291, 294 
Berton, Vic, 142 
Berv, Arthur, 200 
Besson, 128 
Bizet, 143, 266 
Blumel, 173, I94> 2O2 > 22O > 2 37 

Bocal, bassoon, 106 
oboe, 105 

Boehm, 33, 74 

Boieldieu, 291 

Bombard, 90 

Bombardino, 91 

Bombardon, 240 

Bonga, 253 

Bononcini, 116 

Brass, 10 
choir, 27, 28 

Brescia, 40 

Broadwood, 284 

Brown, Tom, 139 

Bruckner, 200 

Buffet, August, Jr, 81, 125 

Bugle, 153, 328 
Kent, 172 
keyed, 172, 173 
scale of, 171 


Biilow, Hans von, 107 
Byrne, T. J., 92, 104 

Caccini, 290 

Cailliet, 142 

Cambert, 97 

Cane, 6, 112 

Carey, George, 270 

Carillon, 263 

Castanet, 253, 266 

Catgut, 10, 266, 271, 275, 291 

Cavaliere, 24, 49, 162, 280 

Cello, 32, 53, 54 

Cembalo, 53 

Cersne, 273 

Cerveny, 109 

Chalumeau, in 

Chaucer, 160, 187 

Chelard, 177 

Cherubim, 67 

Chickering, Jonas, 273 

Chime, 263 

Chladni, 243 

Chopin, 36, 288 

Clagget, 173 

Clarinet, 31, 99, 144, 338 

alto, 127 

bass, 34, 127 

bass, invention of, 127 

complete family of, 1 1 1-34 

contrabass, 128 

invention of, 117 

metal, 132 

mouthpiece, 112, 126, 145, 146 

oscillograph, 352 

technique, 131 

Clarion, 25, 121, 162, 164, 169 
Clarionette, 121 
Clarioun, 160 
Clavichord, 276-78 
Clementi, 286 
Cleopatra, 58 

Conducting the orchestra, 281 
Conn, C. G., 133, 148, 249 
Conservatory system, 101 
Contrabassoon, see Bassoon 
Cor Anglais, 94 
Corelli, 46, 52 
Cornet, 180, 336 

ancestor of, 154, 156 

characteristics of, 154, 183 

complete family of, 151-84 

keyed, 173 

mouthpiece, 182 

scale, 1 66, 174 

-Trumpet feud, 151 

tuning slide, 181 

valve, 174, 176, 177 
Cornetto, 23, 25, 160, 161, 163, 233 
Cornopean, 180 
Corteccia, 23, 162, 212 
Cortesi, 41, 48, 63 
Cousineau, 289 
Cremona, 41 
Cristofori, 273, 282 
Crusaders, 86, 252, 262 
Crwth, 39 
Cymbal, 265, 303 

Debussy, 36, 142 

DeCruck, 142 

Delusse, 100 

Denner, J., 119 

Denner, J, C, 73, 117 

Diston, Henry, 249 

Dorsey, Tommy, 142 

Double reeds, complete family of, 86- 

Drum, 1 68 

bass, 265 

effects, 268 

side, 265 

snare, 265 

trap, 268 

complete family of, 252-70 

Ellington, Duke, 136 
Embouchure, French clarinet, 126 

Italian clarinet, 126 
English horn, see Horn, English 
Erard, S., 284, 290 
"Eroica" symphony, 32, 122 
Euphonium, 242 
"Euridice," 24, 48, 162 

Fagott, 96, 234 
Fagotto, 96, 234 
Ferlandis, 92 
Fiddle, 40 
Fishe, Isaac, 249 
Flageolet, 23, 62 



Florio, P. G., 73 
Fluegelhorn, 156, 183 

characteristics of, 183 
Flute, 337 

alto and bass, 36, 84 

beak, 62 

complete family of, 57-85 

English, 62 

fork fingering, 66 

German, 62 

key, invention of, 65 

metal, 83 

mouthpiece, 72, 146 

open and closed keys, 79, 82 

oscillograph, 350 

ring key, 77 

technique, 85 
Fortunatus, 39 
Frederick the Great, 70, 284 
French horn, see Horn, French 
French kings, 187 
Frequency, determination of pitch, 296- 

Fritz, 119 

Gassner, 124 

Giilet, 101 

Gillette, 142 

Gilmore, 140 

Gittern, 292 

Glockenspiel, 264 

Gluck, 29, 54, 93, 120, 121, 163, 214, 


Goldman, E. F., 270 
Goodman, Benny, 142 
Gordon, Captain, 77 
Gounod, 202 
Gontershausen, von, 124 
Grenadilla wood, 132 
Gresner, 127 
Guadagnini, 41 
Guarnerius, 41 
Guillaume, 232, 234 
Guitar, 292, 295 

double, 293 

technique of, 294 

Hail Bros., 249 
Haliiday, 172 
Hampel, 192, 202 

Handel, 28, 51, 98, 115, 155, 162, 164, 

190, 213, 236, 282 
Harmonics, violin, 54 
Harmony, 19 
Harp, 271, 284 
Irish, 289 
Italian, 288-92 
technique of, 290, 292 
Harpsichord, 21, 23, 25, 27, 31, 48, 276 
leader of orchestra, 280 
technique of, 279 
Haydn, 30, 99, 120, 121, 169, 196, 217, 

265, 267, 282, 287 
Hautboy, 91 

Hearing, threshold of, 2, 3, 302 
Heckel, 95, 108 
Heckelphone, 36, 95 
Heindl, 83 
Helicon, 241, 250 
Helmholtz, 355 

Henry VIII, 62, 161, 187, ail, 254 
Henton, 140 
Hines, Earl, 142 
Hoch, 155, 183 
Hochbrucker, 289 
Hofman, 100 
Horn, ancestor of, 185 
crook, 191 
English, 34, 86, 92 
forest, 190 
French, 344 

complete family of, 185-207 
oscillograph, 353 
hand, 193, 195 
hunting, 188 
keyed, 193 
machine, 192 
mouthpiece, 207 
of Passau, 127 
scale, 205 
stopped, 193, 203 
technique of, 200 
valve, 194, 195 
Hotteterre, Jacques, 73 

Instruments, string vs. wind, 327 
wind, 314 

Jazz, beginning of, 135 

characteristics of, 135, 142 
Jefferson, Thomas, 286, 295 



Kastner, 124 
Kayser, Joe, 138 
Kegas, 1 86 
Keiser, 116, 190 
Kent bugle, 172 
Kerlino, 41 
Kettledrum, 252, 254 
and trumpets, 255 
Keys, why necessary, 337 
Kincaid, 67 
Kithara, 40, 292 
Klose, 124 
Koenig, 55 
Kolbel, 172, 193 
Kruger, 212 
Krupa, 142 

Lanier, Sidney, 83 

Laribee, 83 

LaRocca, 137 

Lefebre, 140, 148 

Legrenzi, 51, 97, 162 

Levy, 155, 183 

Liberati, 155, 183 

Liszt, 36, 288 

Loudness explained, 307-09 

Lully, 23, 51, 96, 162, 189, 254, 282 

Lute, 23,48,49, 5 1 , 53, 271 

Lydian pipes, 58 

Lyre, 53, 271 

Maggini, 41 
Mahler, 261 

Mannheim, Elector of, 53 
Maraca, 253 
Marimba, 264 
Marius, 284 
Massenet, 143 
Mattheson, Johann, 255 
Medici, 48 
Mellophone, 344 

scale, 205 
Mendelssohn, 33, 100, 106, 115, 164, 

197, 219, 236, 282 
Menschel, Hans, 212 
Mersennus, 254 
Meyerbeer, 34, 35, 93, 107, 127, 129, 

143, 177, 198, 236, 257, 291 
Milan, Luis, 293 
Mills, Tom, 269 
Moeremans, 140 

Monteverde, 25, 49, 64, 162, 167, 290 
Moors, 38, 252, 266, 293 
Moussorgsky, 142 

Mouthpiece, clarinet, 1 12, 126, 145, 

cornet, 182 

flute, 72, 146 

French horn, 207 

oboe, 87 

saxhorn, 247 

saxophone, 145 

trumpet, 182 

tuba, 248 

varieties of, 355~5 8 
Mozart, 30, 99, 115, *2I 164, 217, 265, 

277, 285 

Mailer, 175, 194 
Muller, Ivan, 123 
Muris, Jean de, 276 
Music, how it is made, 296-358 

limits of reproduction, 303 

New Orleans, birthplace of jazz, 135 
Nicholson, 75 
Nolan, Frederick, 79 

Oboe, 308, 337 

d'amour, 91 

da caccia, 92 

complete family of, 86-110 

mouthpiece, 87 

oscillograph, 351 
Oliver, King, 137 
Opera, 18, 21 

in English, 342 

Ophicleide, 144, 163, 232, 235, 237 
Orchestra, conducting the, 281 

composition of, n 

how it grew, 1 8 

instrumentation, 12 

seating arrangement, 12 
Orfeo, 25, 49 
Organ, I, 25, 306 
Oscillographs, 347-55 
Overtones explained, 339*55 

Paganini, 42, 54, 294 
Palestrina, 20, 28 
Pallavicmo, 163 
Pepys, 63 ' 
Peregrino, 41 


Percussion, definite pitch, 262 

instruments, 252-70 
importance of, 269 
indefinite pitch, 267 

mallet played, 262 
Peri, 24, 48, 64, 162, 280 
Pfaff, 82 
Pfund, 259 
Philidor, 54 
Piano, 271-88, 314 

frame, 274, 284 

invention of, 282 

technique of, 287 
Piccolo, 32, 64, 84 

introduction into symphony, 218 

violoncello, 51 

violono, 51, S3 
Pipe organ, see Organ 
Pipes of Pan, 57 
Pipes, open and closed, 315, 321, 332, 


Pitch, 300-06 
Pizzicato, violin, 26 
Plato, 58 

Polyphonic music, 28, 31 
Pottag, 199 

Praetorius, 22, 91, 167, 254, 277 
Psaltery, 271 
Ptolemy Auletes, 58 
Purcell, 163 

Quantz, 70 

Rameau, 115, 189 

Range, voice and musical instruments, 


Ravel, 84, 142, 260 

Rebab, 38 

Recorder, 62 

Reeds, 6 

Resonance of wind instruments ex- 
plained, 315-27 

Rinuccini, 48 

Rossini, 34, 80, 93, 198, 258, 294 

Rubinstein, 288 

Ruggeri, 41 

Russel, Peewee, 142 

Sackbut, 211, 215 
Saint-Saens, 143, 197 
Salpinx, 159 

Sarrus, 109 
Sarrusophone, 36, 109 
Sax, Antoine (Adolphe), 34, 127, 143, 
144, 150, 180, 229, 238, 239, 240 
Sax, C. J., 239 

Saxhorn, 180, 182, 232, 240, 247 
Saxophone, 34, 232, 337 
as jazz instrument, 137 
complete family of, 135-50, 147 
craze, 139 
mouthpiece, 145 
production of, 148 
scale, 145 

Saxtromba, 180, 232, 247 
Scale, clarinet, 113, 119 
cornet, 166, 174 
error of valves, 245 
flute, 59 
Greek, 61 
horn, 205 
mellophone, 205 
musical, 2, 3, 302 
natural, 328-34 
saxophone, 145 
trombone, 210, 211, 225 
trumpet, 1 66, 174 
tuba, 243 

Scarlatti, 50, 71, 97, 163, 191 
Schaller, 142 
Schnitzer, 96 
Schubert, 33, 197, 258 
Schumann, 33, 103, 197, 236 
Score, musical, 22 
Segovia, Andres, 294 
Serpent, 163, 232, 235, 238 
Serpentcleide, 235 
Shawm, 87, 96, 101 
Shepherd, R. L., 16 
Shofar, 185 
Silbermann, 284 
Simon, Frank, 270 
Simon, M., 289 

Single reeds, complete family, m-34 
Slides, why necessary, 334 
Smith, Joseph, 274 
Socrates, 58 
Sor, Ferdinand, 293 
Sound, rate of travel, 309-14 
wave, 304 

analyzed, 311-14 
Sousa, 143, 250, 251, 269 



Sousaphone, invention of, 250 

Spinet, 2-79 

Spohr, 291 

Stamitz, 53 

Steffani, 165 

Stehle, 109 

Steinway, 274 

Stolzei, 174, 194 

Storioni, 41 

Stradella, 50, 163 

Stradivarius, 41 

Strauss, Richard, 36, 91, 141, 143 

Stravinsky, 84 

Striggio, 23, 48, 63, 162, 212 

String bass, 32, 54 

choir, 28 

instruments, 271-95 

quartet, 50 

quintet, 50 
Syrinx, 57 

Tambourine, 253, 266 

Tam-tam, 266 

"Tancredi e Clorinda/' 25, 49 

Tapley, 142 

Tartini, 46, 52 

Teagarden, Jack, 142 

Tenoroon, 92 

Theophilus, 272 

Theorbo, 48, 53 

Thomas, Theodore, 199 

Tibia, 61 

Tieffenbriicker, 40 

Tom-tom, 266 

Tone quality explained, 339-55 

Tourte, 10, 46 

Tremolo, violin, 26 

Triangle, 265 

Triebert, loi 

Triebert, Frederic, 102, 108 

Tromba di tirarsi, 216 

Trombone, 32, 33, 217, 334 

as jazz instrument, 223 

bass, 228 

complete family of, 208-31 

introduction into symphony, 218 

Italian, 221 

pedal notes, 220 

scale, 210, 2H, 225 

slide invented, 209 

technique, 222, 224 

tenor vs. bass, 229 

valve, 221 
Tromlitz, 79 
Trompe, 160 
Trumpet, 336 

ancestor of, 152, 156 

and kettledrums, 255 

characteristics of, 153, 183 

complete family of, 151-84 

-Cornet feud, 151 

crook, 167, 177, 179 

keyed, 173 

mouthpiece, 182 

scale, 1 66, 174 

slide, 177, 215 

tuning slide, 181 

used by royalty, 168 

valve, 174, 176, 177 
Trumpette, 160 
Tschaikowsky, 263 
Tuba, 36, 242, 303 

American, 247, 249, 250 

complete family of, 232-51 

mouthpiece, 248 

oscillograph, 354 

over the shoulder, 248 

scale, 243 
valve, 237 
Tuben, 199, 249 
Tympani, first use of three, 257 

machine, 260 

size of kettles, 261 

tuning of, 255 
Tympanist, capabilities of, 258, 261 

Valves, error of, 245 

piston, 34, 173, 194 

rotary, 34, 194 

sub-ventil, 194 

why necessary, 334, 336 
Venuti, 141 
Verdi, 84 
Vereecken, 140 
Vibraphone, 264 
Vibration, rate of, 296-300 
Viola d'amore, 51, 53 

da braccia, 47 

da gamba, 47, 51, 53 
Violin, 7, 327, 342 

bow, 7 

complete family, 38-56 



construction of, 45 

harmonics, 54. 

mute, 54 

pizzicato, 26 

technique, 55 
Violoncello piccolo, 51 
Violino piccolo, 51, 53 
Viols, 23, 25 
Virdung, 254, 279 
Virginal, 279 
Vivaldi, 52 

Waeght, 90 

Wagner, 35, 54, 199, 220, 236, 258, 263 

Waldhorn, 190 

Wallace, William, 236 

Washburn & Godard, 273 
Weber, 35, 123, 198, 266, 294 
Wehner, 83 
Weidinger, 172 
Wendling, 73 
Wieprecht, 237 
Wilson, Teddy, 142 
Wire, steel, 271 
Woodwinds, choir, 27, 28 
materials used, 8 

Xylophone, 264 

Zanetto, 41 

Zinke, 160, 163, 233 

Zither, 293